Juro is currently unavailable
Juro Oravec
Verified Expert in Engineering
Software Developer
London, United Kingdom
Toptal member since March 2, 2022
Bio
Juro's skill set is that of a software startup co-founder—solutions architect, full-stack developer, data engineer, project manager, user researcher, business consultant and strategist, a bit of DevOps, mediocre designer, and a low-key great comedian. He's deployed projects across academia, biotech, pharma, and marketing. Juro excels in highly complex systems.
Portfolio
Self-employed
Rust, Python, Django, Web Development, Software Architecture, PyO3
BenevolentAI
TypeScript, GraphQL, Vue, Vuetify, Apollo, Prisma, D3.js, Scrum, Agile, Node.js...
DTU Biosustain (Novo Nordisk Foundation Center for Biosustainability)
Vue, Vuetify, TypeScript, Python, Flask, Docker, D3.js, Scrum, Git...
Experience
- Vue - 5 years
- Scrum - 5 years
- PostgreSQL - 5 years
- TypeScript - 5 years
- Rapid Prototyping - 5 years
- User Research - 4 years
- Data Visualization - 3 years
- GraphQL - 3 years
Preferred Environment
Visual Studio Code (VS Code), Git, Docker, TypeScript, GraphQL, Vue, Firebase, Amazon Web Services (AWS), CI/CD Pipelines, PostgreSQL
The most amazing...
...project I've worked on is a set of products (web apps) for drug discovery researchers. If such a complex and ambiguous domain can be streamlined, anything can.
Work Experience
Core Maintainer
2025 - PRESENT
Self-employed
- Served as the core maintainer for open source software - a Rust-powered front-end framework for Python.
- Maintained django-components, the Vue/React equivalent for Python and Django. Grew it to 1,500 GitHub stars and around 109,000 monthly downloads across around 70 releases, 460 commits, and 400 merged pull requests.
- Led the project's technical direction by writing detailed design proposals (e.g., multi-version roadmap and major API redesigns), debating them openly, and then building them. Authored 30+ proposals, totaling tens of thousands of words.
- Maintained maturity and trustworthiness through SemVer, CI/CD, tests across 5 Python versions, Windows/Linux, and all major browsers. Ran automatic per-release benchmarks and smart optimizations to support large projects (e.g., infinite depth).
- Rebuilt the rendering engine from the ground up and moved its slowest parts from Python to Rust (PyO3/maturin). Input validation became 6-7x faster. This was followed by a full Rust rewrite, resulting in an additional roughly 3.5x faster render.
- Ensured security and audited numerous times for vulnerabilities. The core was carefully designed for safety (e.g., built a sandboxed Python expression evaluator on top of Ruff's parser—it rewrites user expressions into a safe alternative).
- Designed, discussed, and implemented proposals, triaged and resolved user issues, onboarded new contributors, wrote documentation, and shipped a surrounding ecosystem of add-on packages.
- Collaborated and coordinated with other maintainers in the Django ecosystem to triage issues and improve compatibility.
Technologies: Rust, Python, Django, Web Development, Software Architecture, PyO3
Full-stack Engineer
2020 - 2022
BenevolentAI
- Worked across four web app products that streamlined the drug discovery processes within a team of 15 (including software engineers, UX designers, user researchers, and product managers).
- Scoped and developed an app for visualization and selection of (chemical) compounds in close collaboration with a team of chemoinformaticians with search result volumes in the range of 10,000.
- Developed multiple quick data-driven prototypes in close collaboration with data scientists and UX designs.
- Explored and scoped migration from Vue 2 to Vue 3 in a codebase with about 400 affected files.
- Designed a backward support system for data visualization from another team where the incoming data was versioned (changing over time).
- Integrated various data visualizations, including network graphs, bar and Venn charts, and a molecule 3D visualizer.
- Integrated analytics to a brownfield project and documented best practices of use.
- Led Scrum ceremonies such as retrospectives in a team of 10-15 people.
- Interviewed applicants for software engineering positions (junior to senior).
Technologies: TypeScript, GraphQL, Vue, Vuetify, Apollo, Prisma, D3.js, Scrum, Agile, Node.js, Python, Neo4j, Express.js, Jira, Project Scoping, Git, Visual Studio Code (VS Code), Docker, Amazon Web Services (AWS), CI/CD Pipelines, User Research, REST, PostgreSQL, SQL, Redis, User Authentication, Data Visualization, Material Design, Unix, Rapid Prototyping, Data Privacy, REST APIs, API Integration, JavaScript, APIs, Web Development, Web App Deployment, Back-end, Back-end Development, Front-end, Front-end Development, Full-stack, Full-stack Development, Vanilla JS, HTML, CSS, ECMAScript (ES6), CSS3, HTML5, Jest, Unit Testing, Analytics, Web Analytics, Web Dashboards, Analytical Dashboards, Data-driven Dashboards, Software Development, Event-driven Architecture, Microservices, RESTful Microservices, Microservices Architecture, Webpack, Web Components, Requirements Analysis, Software Engineering, Integration Testing, DOM, HTML DOM, Sass, Dashboards, RDBMS, Mixpanel, Sentry, Linux, STEM, Balsamiq Mockups, Balsamiq, Google Workspace, User Roles, Apollo Server, Webhooks, GraphQL Code Generator, Server Backups, Database Backups, GitLab CI/CD, GitLab, Web Forms, Forms, Data Validation, Amazon S3 (AWS S3), Single-page Applications (SPAs), Telemetry, Ubuntu, Docker Compose, Cache, Authorization, Authentication, ESLint, Puppeteer, Teamwork
Software Engineer, UX Designer, Sales Development Representative, Project Assistant
2019 - 2020
DTU Biosustain (Novo Nordisk Foundation Center for Biosustainability)
- Integrated a web app with an analytics tool (Exponea) and ensured compliance with GDPR and EU cookie law requirements.
- Presented our software at conferences such as ESIB and BIO-Europe, the biggest B2B matchmaking conference on biotech in Europe. Proactively reached out to potential partners and started conversations.
- Conducted the software audit of the web app, finding over 30 errors or suggestions—UX, typos, and code suggestions.
- Contributed to open-source scientific software, collaborating with academics on feature development.
- Did software procurement: searched for, compared, picked, and onboarded colleagues onto a CRM tool that matched their needs.
- Assisted with video content production, organizing meetings, and coordinating people.
- Participated in presenting the final report in Brussels for the Horizon 2020 project.
- Contributed to an academic spin-off: our team worked on a bioinformatics tool, Caffeine, that was part of the EU Horizon 2020 project, DD-DeCaF (Dd-decaf.eu). Our aim was to spin off the Caffeine project into its own.
Technologies: Vue, Vuetify, TypeScript, Python, Flask, Docker, D3.js, Scrum, Git, Visual Studio Code (VS Code), Amazon Web Services (AWS), CI/CD Pipelines, User Research, REST, PostgreSQL, Node.js, SQL, Redis, User Authentication, Data Visualization, Agile, Material Design, Unix, Rapid Prototyping, Sourcing & Procurement, Cookies, General Data Protection Regulation (GDPR), Bloomreach Digital Experience Platform (DXP), B2B, Outreach, Tech Sales, Pitch Decks, Pitch Preparation, Pitch Presentations, UX Audits, UI Audits, IT Audits, Customer Relationship Management (CRM), Open-source Software (OSS), Content Creation, Video Production, Data Privacy, JavaScript, APIs, Web Development, Web App Deployment, Front-end, Front-end Development, Vanilla JS, HTML, CSS, ECMAScript (ES6), CSS3, HTML5, SaaS, Analytics, Web Analytics, Web Dashboards, Analytical Dashboards, Data-driven Dashboards, Software Development, Google Analytics, Microservices, RESTful Microservices, Microservices Architecture, Webpack, Requirements Analysis, Software Engineering, DOM, HTML DOM, Software as a Service (SaaS), Linux, STEM, User Roles, Web Forms, Forms, Single-page Applications (SPAs), Telemetry, Ubuntu, Teamwork
Technology Solutions Consultant
2018 - 2019
Bloomreach
- Collaborated with IT representatives from eCommerce platforms, integrating proprietary marketing automation software and serving as their point-of-contact.
- Served as a tech lead for a pilot implementation project for recommendation systems. With a data scientist, we aimed to identify and deliver use cases for the potential client (CEE-based eCommerce with millions of unique monthly visitors then).
- Standardized and documented the use of CSS and Jinja templates across our team.
- Implemented ads and banners, recommendation features, and A/B testing (using Exponea) for large EU eCommerce players.
- Saved our team time, improved user experience, and enabled less technically-savvy colleagues to create custom web ad banners on their own.
Technologies: JavaScript, jQuery, Jinja, Python, Git, Visual Studio Code (VS Code), Node.js, SQL, Data Privacy, REST APIs, API Integration, APIs, Front-end, Front-end Development, Vanilla JS, HTML, CSS, CSS3, HTML5, SaaS, Analytics, Web Analytics, Software Development, Google Analytics, Technical Consulting, Business Analysis, Requirements Analysis, Software Engineering, DOM, HTML DOM, Software as a Service (SaaS), Dashboards, RDBMS, Webhooks, Web Forms, Forms, Data Validation, Telemetry, HTML Email, Teamwork
Community Coordinator, Project Assistant, Designer
2017 - 2017
Solved
- Had the opportunity to gain insights into the world of startups, sustainability, and human-centered design, as a community coordinator and project assistant.
- Involved in client engagement and improved a client pitch, effectively conveying our services and value proposition.
- Contributed to community coordination and acted as a liaison for experts on the platform.
- Supported projects by attending meetings, taking notes, and handling miscellaneous tasks, ensuring smooth progress and execution.
- Engaged in content development by creating a comprehensive 20-page help guide, contributing to the accessibility of our services and enhancing client understanding.
- Contributed to graphic design and created illustration designs using Adobe Illustrator and Photoshop.
- Engaged in startup insight—administrative, client, and app development meetings.
- Applied human-centered design principles (Ideo.org), focusing on empathy and user experience in developing solutions.
Technologies: Design Thinking, Human-centered Design (HCD), User Research, Projects, Agile Project Management, Sustainability, Graphic Design, Content Creation, Community, Startups, Smart City Technology, Administrative Support, Pitch Decks, SaaS, Software as a Service (SaaS), Teamwork
Experience
Decision-making Algorithm Development and Data Analysis
A university research team had developed a decision-making algorithm that achieves objectivity by computing all scenarios. The research team's clients had requirements of comparing up to 17 attributes. However, their approach could only handle data with up to 5 attributes (columns) due to combinatorial explosion, so 17 was far beyond possible.
I developed a custom algorithm that is faster and can handle greater dimensions. Where the original approach would have taken 10^36 years, my approach took seven days of computation (under the same conditions).
• My approach was bottom-up, leveraging relationships of triplets in pairwise comparison matrices to limit the computational space. To work around the combinatorial explosion, I used statistical sampling for cases with more attributes. Hence, for larger datasets, we needed only a fixed number of random samples (5 million). To ensure correctness, I did both manual and automated audits and visualizations. To further reduce the computing time and costs, I migrated the algorithm to Rust.
Thus, I prepared the groundwork for a data pipeline to enable the research team to offer the algorithm as a software service.
I developed a custom algorithm that is faster and can handle greater dimensions. Where the original approach would have taken 10^36 years, my approach took seven days of computation (under the same conditions).
• My approach was bottom-up, leveraging relationships of triplets in pairwise comparison matrices to limit the computational space. To work around the combinatorial explosion, I used statistical sampling for cases with more attributes. Hence, for larger datasets, we needed only a fixed number of random samples (5 million). To ensure correctness, I did both manual and automated audits and visualizations. To further reduce the computing time and costs, I migrated the algorithm to Rust.
Thus, I prepared the groundwork for a data pipeline to enable the research team to offer the algorithm as a software service.
Full-stack Web App Turn-key Solution
https://github.com/JuroOravec/template-starter-express-typescript-graphql-prisma-vite-vue3With this template, your project should remain maintainable with a team of up to 10-20 people and lower hundreds of database tables/objects.
FEATURES
• Webhooks
• Multi-factor authentication
• Email
• Send emails
• Receive emails (STMP server and SendGrid)
• Email-based webhook
• Deduplication of incoming email
• User authentication
• User session
• Forms
• Validated, typed
• Submission confirmation sent to the submitter
• CI/CD—test, build, and deploy flow in GitLab CI
• Database
• Read/write and version management with Prisma
• Automatic daily/weekly/monthly backups into the Amazon S3 bucket
• Trigger database update and restore through GitLab CI jobs
• Client
• Vuetify 3 UI library
• Server-side rendering (SSR), static site generator (SSG), or single-page application (SPA)
• Smart queue management for system-wide components like modals, snack bars, etc.
• Declarative and fully-typed forms
• Telemetry
• Usage analytics with Mixpanel with fully-typed events
• Error tracking with Sentry
• Both Mixpanel and Sentry tunneled through the server to avoid ad blockers
• Cron jobs
• Used for creating database backups regularly
• Legal
• Consent and cookie management with Termly
FEATURES
• Webhooks
• Multi-factor authentication
• Send emails
• Receive emails (STMP server and SendGrid)
• Email-based webhook
• Deduplication of incoming email
• User authentication
• User session
• Forms
• Validated, typed
• Submission confirmation sent to the submitter
• CI/CD—test, build, and deploy flow in GitLab CI
• Database
• Read/write and version management with Prisma
• Automatic daily/weekly/monthly backups into the Amazon S3 bucket
• Trigger database update and restore through GitLab CI jobs
• Client
• Vuetify 3 UI library
• Server-side rendering (SSR), static site generator (SSG), or single-page application (SPA)
• Smart queue management for system-wide components like modals, snack bars, etc.
• Declarative and fully-typed forms
• Telemetry
• Usage analytics with Mixpanel with fully-typed events
• Error tracking with Sentry
• Both Mixpanel and Sentry tunneled through the server to avoid ad blockers
• Cron jobs
• Used for creating database backups regularly
• Legal
• Consent and cookie management with Termly
Web Scrapers
https://apify.com/jurooravecA rich scraper template that boasts caching, filtering, data transformation, and GDPR compliance out of the box. This template allows for advanced use cases like scraping only new entries. The framework is built on top of the Apify platform with clients' needs in mind. It offers robustness, flexibility, and business needs that many other scrapers on Apify lack.
• Proxy configuration built-in.
• GDPR compliance. Fields, including personal information, are redacted by default, and you must opt-in to include sensitive info.
• Users can define custom data transformation and filtering. No need to use 3rd party services just for that.
• Results can be cached for advanced use cases like scraping only new results.
• Data integrity. How do you ensure that scrapers work correctly as time goes on? Get a report when scraped data changes, compared to a previous run or a "source of truth." The report points out specific fields that changed.
• High configurability. Compare which one allows you to do more: https://apify.com/junglee/amazon-crawler/input-schema vs. https://apify.com/jurooravec/skcris-scraper/input-schema
Examples:
• www.apify.com/jurooravec/skcris-scraper
• www.apify.com/jurooravec/data-change-monitoring
• Proxy configuration built-in.
• GDPR compliance. Fields, including personal information, are redacted by default, and you must opt-in to include sensitive info.
• Users can define custom data transformation and filtering. No need to use 3rd party services just for that.
• Results can be cached for advanced use cases like scraping only new results.
• Data integrity. How do you ensure that scrapers work correctly as time goes on? Get a report when scraped data changes, compared to a previous run or a "source of truth." The report points out specific fields that changed.
• High configurability. Compare which one allows you to do more: https://apify.com/junglee/amazon-crawler/input-schema vs. https://apify.com/jurooravec/skcris-scraper/input-schema
Examples:
• www.apify.com/jurooravec/skcris-scraper
• www.apify.com/jurooravec/data-change-monitoring
90+ Web Scrapers and Data Loaders
Over the years, I scraped data from over 90 websites: science publications, science events, grants, B2B matchmaking events, and other listings.
I never thought much about it until I learned I could monetize scrapers on Apify.com. You can find my best scrapers on www.apify.com/jurooravec.
Examples:
• Found contacts of around 3,000 psychotherapy and psychiatry professionals in Slovakia across several sources
• Public registers from Slovak, French, and the UK (Scrapy)
• Public images from NASA, the European Space Agency, and Solar Dynamics Observatory (Scrapy)
• Public event listings from various biotech websites (Scrapy)
• Reddit (PRAW)
• SoundCloud (custom JavaScript)
• CrunchBase (custom JavaScript)
• Slant.co (custom JavaScript)
I never thought much about it until I learned I could monetize scrapers on Apify.com. You can find my best scrapers on www.apify.com/jurooravec.
Examples:
• Found contacts of around 3,000 psychotherapy and psychiatry professionals in Slovakia across several sources
• Public registers from Slovak, French, and the UK (Scrapy)
• Public images from NASA, the European Space Agency, and Solar Dynamics Observatory (Scrapy)
• Public event listings from various biotech websites (Scrapy)
• Reddit (PRAW)
• SoundCloud (custom JavaScript)
• CrunchBase (custom JavaScript)
• Slant.co (custom JavaScript)
Software Architecture Design for Scraping Data Pipelines and Content Generation (ChatGPT 4)
I designed a data pipeline for a client who needed a solution to:
• Scrape data from multiple sources.
• Surface said data on a map in their web app.
• Generate articles with information from the said data.
The sources included Google Maps, Facebook, TripAdvisor, and several custom scrapers. The data had to be stored in a database that supports geospatial operations. The data had to be extracted regularly, so to ensure data integrity, I developed a monitoring system that could detect unexpected changes in the dataset. The data generation would be done with ChatGPT and the context injection method.
• Scrape data from multiple sources.
• Surface said data on a map in their web app.
• Generate articles with information from the said data.
The sources included Google Maps, Facebook, TripAdvisor, and several custom scrapers. The data had to be stored in a database that supports geospatial operations. The data had to be extracted regularly, so to ensure data integrity, I developed a monitoring system that could detect unexpected changes in the dataset. The data generation would be done with ChatGPT and the context injection method.
Software Architecture Design for "Therapist AI" Conversational Chatbot App
The app aimed to become the "therapist in your pocket." A 3D-rendered "synthetic human" persona would speak to you, and you would reply with your voice. The differentiator of this app was a proprietary intervention methodology. The architecture included a front-end app, text-to-speech and speech-to-text services, persona rendering using SoulMachines, and more.
I designed the high-level software architecture and estimated costs and requirements. To prepare for this job, I completed courses for designing AI chatbots and using AI in healthcare.
I designed the high-level software architecture and estimated costs and requirements. To prepare for this job, I completed courses for designing AI chatbots and using AI in healthcare.
Twitch Streaming Setup: Serverless, Web Apps, Video Generation Service, OBS, Notion Integration
A set of tools to create a Twitch-style video overlay and add interactivity to live streams (videos played based on custom events). The tools used Notion as the source of truth.
Video Generation Service
The service generated a video given a set of inputs. Videos were uploaded to Google Cloud Storage. The Firestore collection was used to store metadata and serve as a cache. Wrote in Python, used Pillow for image, and MoviePy for video generation.
Task Manager
A web app for managing (CRUD) tasks defined in Notion. Modifying tasks scheduled video generation and queued a command to play this video in a video player.
UI Overlay
A web app served as the Twitch-style video overlay. The app rendered UI based on config fetched from the Firestore collection.
Video Player
A simple web app served as a video overlay. It used Firestore as a command queue. The app played a video based on the command when a command arrived.
OBS Panel
A web-based stream deck, conceptually similar to LioranBoard. This allowed controlling OBS remotely.
Web App Back End
GCloud Functions served as back-end orchestration for various web apps.
RTMP Server
Dockerized RTMP server enabling streaming to multiple platforms simultaneously.
This project is unpublished.
Video Generation Service
The service generated a video given a set of inputs. Videos were uploaded to Google Cloud Storage. The Firestore collection was used to store metadata and serve as a cache. Wrote in Python, used Pillow for image, and MoviePy for video generation.
Task Manager
A web app for managing (CRUD) tasks defined in Notion. Modifying tasks scheduled video generation and queued a command to play this video in a video player.
UI Overlay
A web app served as the Twitch-style video overlay. The app rendered UI based on config fetched from the Firestore collection.
Video Player
A simple web app served as a video overlay. It used Firestore as a command queue. The app played a video based on the command when a command arrived.
OBS Panel
A web-based stream deck, conceptually similar to LioranBoard. This allowed controlling OBS remotely.
Web App Back End
GCloud Functions served as back-end orchestration for various web apps.
RTMP Server
Dockerized RTMP server enabling streaming to multiple platforms simultaneously.
This project is unpublished.
Task Mining and Collaborative History App (Web App + Web Extension)
I made a task mining web extension and web app within a tight 2-week timeframe in Oct 2022, which allowed my 25 classmates at Lifbee to compare their literature research strategies.
The web extension saved the browsing history and time spent. The web app displayed the history and visualized it on a graph based on the similarity of the texts of the visited websites.
Later, at a HacknimeTo hackathon (Bratislava, SK, 2022), we pitched this app as a tool for helping high school students to collaborate and learn how to safely collect information on the Internet.
YouTube video: https://www.youtube.com/watch?v=i12jmoJLUX4
The web extension saved the browsing history and time spent. The web app displayed the history and visualized it on a graph based on the similarity of the texts of the visited websites.
Later, at a HacknimeTo hackathon (Bratislava, SK, 2022), we pitched this app as a tool for helping high school students to collaborate and learn how to safely collect information on the Internet.
YouTube video: https://www.youtube.com/watch?v=i12jmoJLUX4
Web App: Spotify-Strava Integration (Spotify Playlists Based on Strava Activities)
Integration between Spotify and Strava created Spotify playlists from songs you listened to during your Strava activity. The playlist was also added to the Strava activity description.
User Flow
User logs in with email or social login and connects their Strava and Spotify accounts. The dashboard displays all created Spotify playlists. The user configures how to format a new playlist and whether and how to update the Strava activity description.
Back End
Written in TypeScript, Node.js, and Express.js. I used REST API and Passport.js to handle user authentication.
For each user, the app polled Spotify for recently played songs with their access token and cached the songs in the PostgreSQL database. Likewise, for each user, a webhook was registered on Strava to listen for the "activity created" event. On the event, the songs were retrieved from the cache, and a new Spotify playlist was created with them.
Front End
Vue 2, Vuetify
Deployment
The back end was deployed on AWS EC2, the front end on S3, and the database on RDS.
The project was stopped because of poor integration support. Strava webhooks can have delays up to 30 minutes, and Spotify was rate-limiting requests per app, not users, so this was not scalable.
User Flow
User logs in with email or social login and connects their Strava and Spotify accounts. The dashboard displays all created Spotify playlists. The user configures how to format a new playlist and whether and how to update the Strava activity description.
Back End
Written in TypeScript, Node.js, and Express.js. I used REST API and Passport.js to handle user authentication.
For each user, the app polled Spotify for recently played songs with their access token and cached the songs in the PostgreSQL database. Likewise, for each user, a webhook was registered on Strava to listen for the "activity created" event. On the event, the songs were retrieved from the cache, and a new Spotify playlist was created with them.
Front End
Vue 2, Vuetify
Deployment
The back end was deployed on AWS EC2, the front end on S3, and the database on RDS.
The project was stopped because of poor integration support. Strava webhooks can have delays up to 30 minutes, and Spotify was rate-limiting requests per app, not users, so this was not scalable.
Web App: Calorie Counter Demo
https://github.com/JuroOravec/calorie-counter-demo-2022A demo project showcasing the best practices for a TypeScript-based web app.
Features
• CRUD resources (e.g., user can create food entries).
• Authentication (email and password, social login) using authentication cookies.
• User roles (user, administrator).
• Tiered GraphQL queries (different user roles have access to different queries).
• User settings
• Administrator reports and views.
Front End
• Apollo GraphQL
• TypeScript
• Vue.js 3
• Vuetify
Back End
• GraphQL server
• TypeScript
• Prisma with PostgreSQL.
• Authentication layer with Express.js and Passport.js.
• Redis as a cache store.
• Dockerized
• Reverse proxy using Nginx.
Demo: https://www.youtube.com/watch?v=bidlVoby72Y
Features
• CRUD resources (e.g., user can create food entries).
• Authentication (email and password, social login) using authentication cookies.
• User roles (user, administrator).
• Tiered GraphQL queries (different user roles have access to different queries).
• User settings
• Administrator reports and views.
Front End
• Apollo GraphQL
• TypeScript
• Vue.js 3
• Vuetify
Back End
• GraphQL server
• TypeScript
• Prisma with PostgreSQL.
• Authentication layer with Express.js and Passport.js.
• Redis as a cache store.
• Dockerized
• Reverse proxy using Nginx.
Demo: https://www.youtube.com/watch?v=bidlVoby72Y
Static Site: Personal Site
A personal (blog) static site. A static site is a website that is pre-built. It does not require a server running when published. The website posts are written in Markdown.
Implementation
Implemented using Gridsome (Vue.js part of Gatsby). Compared to a regular Gridsome project, this is a heavily customized setup and was the first of its kind to support GraphQL code generation and offers TypeScript support unlike any other. This project required fixing issues across three other open-source packages.
Website Features
• Search across all posts.
• Recommended and related posts.
• Git-based comments using Vssue.
• Analytics (Mixpanel, Sentry).
• SEO features such as meta tags (including Open Graph and Twitter cards(, sitemap, and robots.txt.
• RSS feed
• Pre-compression of static files.
Codebase Features
• Full TypeScript support.
• Fine control over the data flow 1) Validation of input data from Markdown files, 2) Enrichment of the internal data, 3) GraphQL schema with custom resolvers.
• Vue.js components can be embedded inside Markdown files.
• Vuetify support
• GraphQL codegeneration (autogenerated TypeScript files based on GraphQL queries).
• Modularized project structure.
Deployment
Netlify
Implementation
Implemented using Gridsome (Vue.js part of Gatsby). Compared to a regular Gridsome project, this is a heavily customized setup and was the first of its kind to support GraphQL code generation and offers TypeScript support unlike any other. This project required fixing issues across three other open-source packages.
Website Features
• Search across all posts.
• Recommended and related posts.
• Git-based comments using Vssue.
• Analytics (Mixpanel, Sentry).
• SEO features such as meta tags (including Open Graph and Twitter cards(, sitemap, and robots.txt.
• RSS feed
• Pre-compression of static files.
Codebase Features
• Full TypeScript support.
• Fine control over the data flow 1) Validation of input data from Markdown files, 2) Enrichment of the internal data, 3) GraphQL schema with custom resolvers.
• Vue.js components can be embedded inside Markdown files.
• Vuetify support
• GraphQL codegeneration (autogenerated TypeScript files based on GraphQL queries).
• Modularized project structure.
Deployment
Netlify
API Reverse Engineering - Clubhouse
When the Clubhouse was still fresh and only on iOS, I used Proxyman to explore and document the Clubhouse API interface and the used domain models. A jailbroken iPhone was required to work around the certificate pinning.
This project is unpublished.
This project is unpublished.
Web Extension: Reddit Feed Filtering (Frequency-based)
https://github.com/JuroOravec/filter-reddit-feedOVERVIEW
Filtering posts on social media is too simplistic for my needs. I want to see less of something, but I still want to allow for serendipity. I made a web extension that filters out only a fraction of Reddit posts. The web extension provides a pop-up UI to configure the frequency at which you see posts (frequency is set per subreddit).
IMPLEMENTATION
I used a web extension API to read the user settings and web page (on the reddit.com domain). Then, it reads, parses (and listens for future changes to) the webpage's HTML. Based on the frequency configured by the user, posts are hidden based on their parsed subreddit. Posts are hidden by removing them from the HTML. User settings are persisted per user browser profile via a web extension API.
The UI is implemented in React and TypeScript.
Filtering posts on social media is too simplistic for my needs. I want to see less of something, but I still want to allow for serendipity. I made a web extension that filters out only a fraction of Reddit posts. The web extension provides a pop-up UI to configure the frequency at which you see posts (frequency is set per subreddit).
IMPLEMENTATION
I used a web extension API to read the user settings and web page (on the reddit.com domain). Then, it reads, parses (and listens for future changes to) the webpage's HTML. Based on the frequency configured by the user, posts are hidden based on their parsed subreddit. Posts are hidden by removing them from the HTML. User settings are persisted per user browser profile via a web extension API.
The UI is implemented in React and TypeScript.
Web Extension: Click Censor
https://www.youtube.com/watch?v=z4hu78LpvVEA web extension that added a simple utility; when you selected a text and right-clicked it, you had the option to "censor" that text. This action would search for all occurrences of the phrase in the current webpage and hide them.
IMPLEMENTATION
The web extensions were written in JavaScript. Interaction with the page was handled using web extension browser API. The project used a queue to handle user actions asynchronously.
Currently, the project is unlisted from Chrome Webstore as the policies changed, and the web extension was using Sentry for error detection.
IMPLEMENTATION
The web extensions were written in JavaScript. Interaction with the page was handled using web extension browser API. The project used a queue to handle user actions asynchronously.
Currently, the project is unlisted from Chrome Webstore as the policies changed, and the web extension was using Sentry for error detection.
Lodash-like Library Defined Inside Jinja2
https://github.com/JuroOravec/powerjinjaWhile working at Exponea (now part of Bloomreach), in my spare time, I was looking for ways to simplify common tasks in my team, like writing web banners. For this, we used Jinja2, a Python templating engine.
Using macros, I initially developed a library that can pipe functions, allowing to call functions one after another within the Jinja2 environment. This led to more streamlined coding processes and enhanced productivity.
Next, I wanted to take it a step further and solve another common problem: data integrity. I had to define a lot more functions for this to work. Inspired by the popular JavaScript utility library, Lodash, these functions were defined separately. In the end, this Lodash-like library consisted of 260 functions defined purely in Jinja2.
Using macros, I initially developed a library that can pipe functions, allowing to call functions one after another within the Jinja2 environment. This led to more streamlined coding processes and enhanced productivity.
Next, I wanted to take it a step further and solve another common problem: data integrity. I had to define a lot more functions for this to work. Inspired by the popular JavaScript utility library, Lodash, these functions were defined separately. In the end, this Lodash-like library consisted of 260 functions defined purely in Jinja2.
Tool for Synchronising Text Translations Across a Vue Codebase
PROBLEM
Vue-i18n is a popular Vue library for supporting multiple languages on a website (internationalization, aka i18n).
In such Vue projects, some software developers prefer when text translations are co-located with the rest of their code. However, when the text needs to be translated, translators require a single file with all the tokens needed.
Hence, this leads to a question of where to store the information—either a single file (easy to translate) or across many files (easy to use in code). Both have their trade-offs, which opens up human errors because either way, one party must update their files manually.
SOLUTION
A software library that syncs and automatically converts between the two styles of managing i18n translations. The library can:
• Extract all translations from a Vue codebase into a single JSON file.
• Update the translations in the codebase with the new values in the single JSON.
• Check for discrepancies between the codebase and JSON, thus ensuring conformity.
Vue-i18n is a popular Vue library for supporting multiple languages on a website (internationalization, aka i18n).
In such Vue projects, some software developers prefer when text translations are co-located with the rest of their code. However, when the text needs to be translated, translators require a single file with all the tokens needed.
Hence, this leads to a question of where to store the information—either a single file (easy to translate) or across many files (easy to use in code). Both have their trade-offs, which opens up human errors because either way, one party must update their files manually.
SOLUTION
A software library that syncs and automatically converts between the two styles of managing i18n translations. The library can:
• Extract all translations from a Vue codebase into a single JSON file.
• Update the translations in the codebase with the new values in the single JSON.
• Check for discrepancies between the codebase and JSON, thus ensuring conformity.
Tool for Modifying Complex 3rd-party Vue Components
PROBLEM
In Vue projects, developers can use visual components from 3rd parties. This usually saves time. Usually, such components couple logic and presentation (e.g., a calendar component knows how to select a time range).
However, a problem arises when you need to modify the 3rd-party components because of the coupling. You have only a few ways of changing the 3rd-party components (component inputs and CSS selectors). If you need modifications the authors didn't anticipate, you may need to re-implement the whole 3rd-party component, which costs extra time and effort.
SOLUTION
Vue stub is a software library that allows replacing otherwise inaccessible parts of 3rd-party Vue components. This unlocks a new way developers can modify 3rd-party libraries and is useful, especially for unit testing.
In Vue projects, developers can use visual components from 3rd parties. This usually saves time. Usually, such components couple logic and presentation (e.g., a calendar component knows how to select a time range).
However, a problem arises when you need to modify the 3rd-party components because of the coupling. You have only a few ways of changing the 3rd-party components (component inputs and CSS selectors). If you need modifications the authors didn't anticipate, you may need to re-implement the whole 3rd-party component, which costs extra time and effort.
SOLUTION
Vue stub is a software library that allows replacing otherwise inaccessible parts of 3rd-party Vue components. This unlocks a new way developers can modify 3rd-party libraries and is useful, especially for unit testing.
Workflow and Tool for Managing User Research Questions as Mind Maps (Assumption Map)
https://www.youtube.com/watch?v=lefayQzdvFgAs a user researcher or designer, your go-to tool or interface may be a whiteboard, not a spreadsheet. However, spreadsheets and structured data are integral for effectively managing user research activities.
What would it look like if visual thinking and whiteboards were THE sources of truth? I designed a workflow and developed a software tool that allows just that.
The software tool can convert mind map diagrams (Draw.io) into CSVs (spreadsheets) and back.
Hence, designers can use a visual mind map to generate assumptions to check during user interviews. Once they get feedback from users, they can update the mind map by simply uploading the updated spreadsheet.
What would it look like if visual thinking and whiteboards were THE sources of truth? I designed a workflow and developed a software tool that allows just that.
The software tool can convert mind map diagrams (Draw.io) into CSVs (spreadsheets) and back.
Hence, designers can use a visual mind map to generate assumptions to check during user interviews. Once they get feedback from users, they can update the mind map by simply uploading the updated spreadsheet.
[Biotech] Product and Business Development | Lifbee Bioentrepreneurship Program
Lifbee is a 7-month bioentrepreneurship program. Participants worked in teams during the last four months on their startup ideas. The ideas were validated, and business models were developed and pitched to investors and industry partners. Our multidisciplinary team worked on a bio-based foam material for the packaging industry.
ACTIVITIES
• Business development. Developed the business plan, pursued partnerships with academia and industry, and prepared and presented pitch decks. Conducted sales outreach to potential clients and stakeholders.
• Financial planning. Developed financial models to analyze market size, estimate operational and startup costs, assess profitability, and calculate the break-even point.
• Product development. Performed user research and market analysis, utilizing scrapers to collect market data. Developed the product roadmap.
• Research and development (R&D). Conducted literature review.
• Marketing. Conducted keyword analysis, created a landing page with analytics, and ran targeted advertising campaigns through Google Ads to assess customer acquisition costs (CAC). Used Figma and Gridsom, leveraging ChatGPT and Midjourney to generate content and imagery for the website.
• Project management with Jira.
ACTIVITIES
• Business development. Developed the business plan, pursued partnerships with academia and industry, and prepared and presented pitch decks. Conducted sales outreach to potential clients and stakeholders.
• Financial planning. Developed financial models to analyze market size, estimate operational and startup costs, assess profitability, and calculate the break-even point.
• Product development. Performed user research and market analysis, utilizing scrapers to collect market data. Developed the product roadmap.
• Research and development (R&D). Conducted literature review.
• Marketing. Conducted keyword analysis, created a landing page with analytics, and ran targeted advertising campaigns through Google Ads to assess customer acquisition costs (CAC). Used Figma and Gridsom, leveraging ChatGPT and Midjourney to generate content and imagery for the website.
• Project management with Jira.
[Biotech] Feasibility Study Preparation and Literature Review
After the Lifbee bioentrepreneurship program, I established the groundwork for funding and evaluating the feasibility of the biotech startup idea (cellulose foam for packaging).
ACTIVITIES
• Grant proposal writing. Prepared and submitted grant proposals.
• Literature review. Designed an in-depth analysis to track 65 parameters across 1,000 research papers to understand the process of manufacturing cellulose foam (link to the spreadsheet: https://docs.google.com/spreadsheets/d/1zi1DXLgbPNxBtGxBsKnWrUIOhlGW9fCKt_yyhOsUdno/).
• Identification of key opinion leaders (KOLs). Mapped and connected with leading experts in the field of cellulose foam technology (link to the spreadsheet: https://docs.google.com/spreadsheets/d/1-z4aPlGv3dO6eJfS_8rmEC-a3OZuxNYnsiwkUv4wwaU/).
• Lead generation. Assembled a dataset of Slovak research groups working in related areas.
• Web scraping. Wrote scrapers to extract metadata from most major scientific publisher platforms.
• Project management. Defined goals, timelines, and key milestones.
ACTIVITIES
• Grant proposal writing. Prepared and submitted grant proposals.
• Literature review. Designed an in-depth analysis to track 65 parameters across 1,000 research papers to understand the process of manufacturing cellulose foam (link to the spreadsheet: https://docs.google.com/spreadsheets/d/1zi1DXLgbPNxBtGxBsKnWrUIOhlGW9fCKt_yyhOsUdno/).
• Identification of key opinion leaders (KOLs). Mapped and connected with leading experts in the field of cellulose foam technology (link to the spreadsheet: https://docs.google.com/spreadsheets/d/1-z4aPlGv3dO6eJfS_8rmEC-a3OZuxNYnsiwkUv4wwaU/).
• Lead generation. Assembled a dataset of Slovak research groups working in related areas.
• Web scraping. Wrote scrapers to extract metadata from most major scientific publisher platforms.
• Project management. Defined goals, timelines, and key milestones.
[Biotech] Market Research on a Nutritional Product for a Biotech R&D Company
I conducted a literature review and market research for a Slovak biotech R&D company operating in biofuels and corn derivatives. The aim was to assess the feasibility and inform on financials. Results were presented to the client.
I used web scraping to estimate the bid/ask prices using data from a chemical procurement marketplace website (Guidechem).
I used web scraping to estimate the bid/ask prices using data from a chemical procurement marketplace website (Guidechem).
[Biotech] User Researcher, UI/UX | Bioinformatics Tool for Biologists
In 2019, my last year of university (BSc Biotechnology at the University of Edinburgh), I worked on designing and carrying out genetic modifications on baker's yeast.
It was then that I recognized that the day-to-day work of biologists could be made simpler if there was a robust system for designing genetic modifications that could predict the outcomes.
ACTIVITIES
• User research interviews. Conducted comprehensive interviews with academics (UoE) and industry professionals (Ingenza) to understand the needs and challenges.
• UI/UX design. Developed mockup designs and a feature list for a potential bioinformatics tool, translating the gathered insights into a tangible concept.
It was then that I recognized that the day-to-day work of biologists could be made simpler if there was a robust system for designing genetic modifications that could predict the outcomes.
ACTIVITIES
• User research interviews. Conducted comprehensive interviews with academics (UoE) and industry professionals (Ingenza) to understand the needs and challenges.
• UI/UX design. Developed mockup designs and a feature list for a potential bioinformatics tool, translating the gathered insights into a tangible concept.
[Biotech] Honors Project: Genetic Engineering of Baker's Yeast | Wet Lab and Bioinformatics
Grade: First Class
Honors project: Expression of a metabolic pathway for the production of methacrylate esters in yeast.
ACTIVITIES
• Computational methods: metabolic modeling, bioinformatics, linear programming, cobrapy, Python
• Lab work: metabolic engineering
• Non-biological courses taken: several variable calculus, fundamentals of entrepreneurship, programming, French language
Honors project: Expression of a metabolic pathway for the production of methacrylate esters in yeast.
ACTIVITIES
• Computational methods: metabolic modeling, bioinformatics, linear programming, cobrapy, Python
• Lab work: metabolic engineering
• Non-biological courses taken: several variable calculus, fundamentals of entrepreneurship, programming, French language
Native Ads Strategy Consulting Project for News Agency
Lipa SC was a student organization specialized in connecting students with real-life experience to work on consulting cases.
I worked as a team lead and led a dynamic team of four. Our focus was on providing consultation to a notable Slovak newspaper agency, specifically advising on native advertising strategies.
RESPONSIBILITIES
• Handled project management.
• Performed market research analysis, including trends, audience behavior, and preferences.
• Conducted competitor analysis and identified unique selling points and areas of opportunity for our clients to distinguish themselves in the market
• Did client pitching.
• Collaborated, mentored, facilitated communication, and provided guidance to ensure the successful execution of the project.
I worked as a team lead and led a dynamic team of four. Our focus was on providing consultation to a notable Slovak newspaper agency, specifically advising on native advertising strategies.
RESPONSIBILITIES
• Handled project management.
• Performed market research analysis, including trends, audience behavior, and preferences.
• Conducted competitor analysis and identified unique selling points and areas of opportunity for our clients to distinguish themselves in the market
• Did client pitching.
• Collaborated, mentored, facilitated communication, and provided guidance to ensure the successful execution of the project.
Dropshipping eCommerce: Selling Prints of Space Images
In 2017, I built a back end for a dropshipping business selling prints of space images available in the public domain (NASA, ESA, etc.).
ACTIVITIES
• Web scraping. Scoured NASA, ESA, and other websites to gather space images under the public domain, ensuring compliance with copyright laws.
• Web application development. Built an internal web app using JavaScript, JQuery, and MySQL that allowed us to pick suitable images for selling by rating them. This streamlined our selection process and enhanced our efficiency.
• Integration with printing services. Successfully integrated with Pwinty.com, a system that offers printing jobs, facilitating a seamless process from image selection to print production.
ACTIVITIES
• Web scraping. Scoured NASA, ESA, and other websites to gather space images under the public domain, ensuring compliance with copyright laws.
• Web application development. Built an internal web app using JavaScript, JQuery, and MySQL that allowed us to pick suitable images for selling by rating them. This streamlined our selection process and enhanced our efficiency.
• Integration with printing services. Successfully integrated with Pwinty.com, a system that offers printing jobs, facilitating a seamless process from image selection to print production.
[Media] Videographer and President of Student Filmmaking Society
I produced over 70 videos, including commissioned projects, competitions, and YouTube content (25,000 views). Videos ranged from creative storytelling to tailored brand content, such as a series of 24 workout videos for a fitness brand. My work also included managing a team of 4 and coordinating on-screen performances of up to 30 people.
As part of a team of 6, we planned events, competitions, sponsorships, and rentals.
As part of a team of 6, we planned events, competitions, sponsorships, and rentals.
Education
2015 - 2019
Bachelor's Degree in Biotechnology
University of Edinburgh - Edinburgh, UK
Certifications
JUNE 2023 - PRESENT
Venture Capital (VC) 3-month Program
Zero One Hundred Academy
FEBRUARY 2023 - PRESENT
Lifbee 7-month Bioentrepreneurship Program
Lifbee
DECEMBER 2022 - PRESENT
Conversation AI Courses
Conversation Design Institute
DECEMBER 2022 - PRESENT
Conversational AI Courses
Soul Machines Academy
NOVEMBER 2022 - PRESENT
HelloAI Professional
EIT Health
MARCH 2017 - PRESENT
3 Day Startup (3DS) Entrepreneurship Bootcamp
3 Day Startup
OCTOBER 2016 - PRESENT
Personal Assistant (Carer) Training
Mochridhe Ltd
Skills
Libraries/APIs
Vue, Node.js, D3.js, jQuery, REST APIs, Spotify API, Passport.js, Puppeteer, React, CheerioJS, Python Imaging Library, Stripe, Stripe API, Facebook Open Graph API, Lodash, NumPy, SymPy
Tools
Git, Prisma, Sentry, NGINX, GitLab CI/CD, Docker Compose, GitHub, Spreadsheets, Webpack, Adobe CC, Jira, Open Broadcaster Studio (OBS), Cron, GitLab, SendGrid, Google Analytics, Balsamiq, Balsamiq Mockups, Stripe Checkout, ChatGPT, Figma, Draw.io, Mailchimp, Google Workspace
Languages
TypeScript, GraphQL, Python, SQL, JavaScript, HTML, CSS, ECMAScript (ES6), CSS3, HTML5, Sass, XML, Python 3, Rust
Frameworks
Vuetify, Express.js, Jinja, Chrome, Vanilla JS, Jest, Flask, Scrapy, Nuxt.js, Flutter, Django
Paradigms
REST, Agile, Scrum, Design Thinking, Rapid Prototyping, Unit Testing, Event-driven Architecture, Microservices, Microservices Architecture, Requirements Analysis, HTML DOM, Serverless Architecture, Application Architecture, Web Architecture, Real-time Messaging Protocol (RTMP), Automation, ETL, Agile Project Management, Human-centered Design (HCD), UI Design, UX Design, Web App Design, Linear Programming, B2B, Web UI Design, Functional Programming
Platforms
Docker, Firebase, Mixpanel, Netlify, Visual Studio Code (VS Code), Ubuntu, Linux, AWS Lambda, Amazon Web Services (AWS), Unix, DigitalOcean, AWS Elastic Beanstalk, Google Cloud Platform (GCP), X (formerly Twitter), iOS, Anaconda, Google Ads, Heroku
Storage
PostgreSQL, Redis, RDBMS, Database Backups, Web Forms, Data Validation, Amazon S3 (AWS S3), Cache, Neo4j, MongoDB, Google Cloud, Cloud Firestore, Google Cloud Storage, Data Integration, PostGIS
Other
CI/CD Pipelines, User Research, Web Extensions, User Authentication, Web Scraping, Data Visualization, Material Design, Biology, Apollo, Project Scoping, API Integration, User Roles, Apollo Server, Webhooks, APIs, GraphQL Code Generator, Server Backups, Forms, Single-page Applications (SPAs), Telemetry, Authorization, Authentication, ESLint, Apify SDK, HTML Parsing, Data Extraction, Data Collection, Team Leadership, Leadership, Web Development, Web App Deployment, Chrome Extensions, Data Scraping, Data Privacy, Data Processing, Scraping, Back-end, Back-end Development, Front-end, Front-end Development, Full-stack, Full-stack Development, SaaS, Analytics, Web Analytics, Web Dashboards, Analytical Dashboards, Data-driven Dashboards, Software Development, Architecture, RESTful Microservices, STEM, Technical Consulting, Business Analysis, Software Engineering, DOM, Software as a Service (SaaS), Lambda Functions, Dashboards, Gatsby, Google SEO, SEO Tools, RSS Feeds, Sitemaps, Serverless, Google Cloud Functions, Open Broadcaster Software (OBS), Mail Servers, SSL Certificates, Session Handling, Multiple Factor Analysis (MFA), Server-side Rendering (SSR), Payment Gateways, UFW, HTML Email, Spreadsheet Automation, Geographic Information Systems, OpenAI GPT-4 API, OpenAI GPT-3 API, Web Components, Integration Testing, Video Streaming, Broadcasting, Video Livestreaming, Vite, Recommendation Systems, Comments, Message Queues, Reverse Engineering, Mobile App Development, UI Automation, Consulting, Market Research & Analysis, Competitor Analysis & Profiling, Pitch Decks, Pitch Preparation, Team Mentoring, Communication, Time Management, Projects, Sustainability, Graphic Design, Content Creation, Community, Startups, Smart City Technology, Administrative Support, Collaboration, Interdisciplinary Collaboration, UX Research, UI Research, Research, Life Science, Teamwork, Technical Writing, Business Development, Entrepreneurship, IT Project Management, eCommerce, Supply Chain, Ideation, Prototyping, B2C Marketing, Complex Problem Solving, Optimization, Software, Assistance, Calculus, Internalization, i18n, Abstract Syntax Trees (AST), Software Architecture, Conversational Interfaces, Chatbot Conversation Design, Speech-to-Text (STT), Text-to-Speech (TTS), Agile Sprints, Data Integrity Testing, Text Generation, Lead Generation, B2B Lead Generation, Sourcing & Procurement, Cookies, General Data Protection Regulation (GDPR), Bloomreach Digital Experience Platform (DXP), Outreach, Tech Sales, Pitch Presentations, UX Audits, UI Audits, IT Audits, Customer Relationship Management (CRM), Open-source Software (OSS), Video Production, Investing, Recruiting, Contract, Customer Service, Contract Negotiation, Venture Capital, Venture Funding, New Business Ventures, Funding Strategy, Sourcing, Investment Funds, Impact Investing, Equity Investment, Due Diligence, Deal Structuring, Portfolio Management, Financial Modeling, Financial Planning & Analysis (FP&A), Literature Review, Business Planning, Business Model Canvas, Business Models, B2B Partnerships, Strategic Partnerships, Strategy Development, Startup Growth Strategy Development, Budgeting, Product Roadmaps, Keyword Analysis, Landing Pages, Landing Page Optimization, Landing Page Design, Web Marketing, SEO Marketing, Google Advertising, Paid Advertising, Grant Proposals, Grants, Science, Biotechnology, Data Analysis, Lean Project Management, Feasibility Studies, Strategic Planning, Stakeholder Engagement, Stakeholder Interviews, Hackathons, Browsers, Event Tracking, Tracking, Web Applications, User Experience (UX), Quantitative User Research, CSV, File Conversion, Filmmaking, Short Films, Videos, Videography, Events, Event Management, Email Mailing Lists, Conversational UX, Conversational UI, Conversational Design, Chatbots, Artificial Intelligence (AI), Healthcare IT, Healthcare Services, Medical Imaging, Matrix Algebra, Algorithms, PyO3
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring