Robert is available for hire
Hire RobertRobert Gevorgyan
Verified Expert in Engineering
JavaScript Developer
Dublin, Ireland
Toptal member since February 9, 2021
Bio
Robert is an engineer with over a decade of experience. His academic background in computer science and builder mindset make him a great fit for product engineering and AI roles. Detail-oriented and quality-driven, he delivers exceptional products. Robert worked closely with several former Google, Microsoft, and Meta founders. Trusted as a founding and staff-level engineer across multiple engagements, he thrives in solving complex problems and takes full ownership of the projects he leads.
Portfolio
compassivf.com
Artificial Intelligence (AI), Large Language Models (LLMs), Full-stack...
Hearst - Technology
Python, AI Agents, Agentic AI, Claude Code, Google Cloud Platform (GCP)...
Tiny Co Holdings, Inc
APIs, TypeScript, Front-end, AI Agents, Back-end, Node.js, Architecture...
Experience
- Full-stack - 10 years
- JavaScript - 10 years
- Node.js - 9 years
- React - 9 years
- TypeScript - 7 years
- PostgreSQL - 7 years
- Cloud Architecture - 5 years
- Artificial Intelligence (AI) - 3 years
Work Experience
Full-stack AI Agents/LLM Engineer
2026 - 2026
compassivf.com
- Developed Hope, an empathetic AI companion to support the clients through their IVF journey.
- Architected the system from scratch with minimal maintenance costs.
- Built fully using parallel, fully agentic AI workflows.
Technologies: Artificial Intelligence (AI), Large Language Models (LLMs), Full-stack, Web Development, Claude Code, Gemini API, Vibe Coding, AI Agents, PostgreSQL, Pgvector, Cloud Architecture, TypeScript, LangGraph, LangChain, ElevenLabs Solutions, WebSockets, CI/CD Pipelines, SSH, Solution Architecture
AI-first Full-stack Developer
2026 - 2026
Hearst - Technology
- Developed a pipeline to evaluate AI and OCR models for handwritten and printed documents.
- Built a pipeline to automatically sort and extract data using AI/OCR from scanned physical mail.
- Contributed to a product for magazine editors to leverage AI in an Adobe InDesign plugin.
Technologies: Python, AI Agents, Agentic AI, Claude Code, Google Cloud Platform (GCP), TypeScript, Next.js, Optical Character Recognition (OCR)
Full-stack LLM/TypeScript Staff Engineer
2025 - 2025
Tiny Co Holdings, Inc
- Engineered and delivered a full automated Agent-to-Agent integration using LangChain/Langgraph and Cursor APIs.
- Enhanced the AI agent knowledge graph using AI design patterns such as Human in the Loop, Suggested Actions, Parallel Execution, etc.
- Created a smooth multi-tenant user onboarding experience with several API integrations such as Google, GitHub App, and more.
Technologies: APIs, TypeScript, Front-end, AI Agents, Back-end, Node.js, Architecture, Full-stack, Large Language Models (LLMs), SvelteKit, Svelte, LangChain, Rust, LangGraph, Drizzle, Claude, Solution Architecture, AI Programming, Agentic AI, Agentic Frameworks, Software Architecture, System Design, Artificial Intelligence (AI), Python, Large Language Model Operations (LLMOps), Zod, Zod Schema
React/Python/Next.js/GCP Full-stack Engineer
2025 - 2025
Greyball, Inc
- Led the architecture and development of the web application, resulting in a successful demo and first client signing, significantly increasing ARR.
- Enhanced the RAG pipeline document intake and storage and introduced streaming, leading to lower latency and better AI responses.
- Contributed to the agentic AI workflow design, introducing legal AI agents for e2e AI-assisted work.
- Oversaw the distributed system migration from multiple services to GCP, ensuring a unified platform experience.
Technologies: JavaScript, React, HTML, XML, CSS, DocAve, Google Cloud, PostgreSQL, Node.js, Supabase, Supabase Auth, Artificial Intelligence (AI), Retrieval-augmented Generation (RAG), Google Cloud Platform (GCP), Python, Architecture, Databases, Large Language Models (LLMs), OpenAI API, Anthropic, Google API, Drizzle, LangChain, LangGraph, Shadcn, Claude, Solution Architecture, AI Programming, Agentic AI, Agentic Frameworks, Software Architecture, System Design, AI Agents, Large Language Model Operations (LLMOps), Zod, Zod Schema
Senior Developer
2023 - 2025
Schlafender Hase GmbH
- Ported a desktop application to the web, ensuring smooth releases and seamless migration of 30+ users.
- Oversaw the team to ensure proper architecture and development of features such as document layout, deep zooming, and canvas overlay annotations.
- Integrated parts of the desktop application directly from C++ using WASM.
- Optimized the rendering for large projects, cutting the first meaningful render time by more than 10 times.
- Integrated Open AI API to spot differences in images and text documents.
Technologies: React, TypeScript, HTML5, CSS, HTML Canvas, JavaScript, Canvas, SVG, Jest, Next.js, OpenSeadragon, WebAssembly (Wasm), React Query, CI/CD Pipelines, Agile, REST, Sentry, Architecture, Scalability, Cloud Architecture, Web Applications, Content Delivery Networks (CDN), Vercel, Retrieval-augmented Generation (RAG), Databases, Redis, Dashboards, Web UI, Web UX, TanStack Router, Minimum Viable Product (MVP), Startups, GitHub, OpenAI, Cursor AI, Tailwind CSS, OpenAI API, Webpack, React Testing Library, API Integration, ChatGPT, Software as a Service (SaaS), Responsive UI, Responsive Web Design (RWD), Zustand, ECMAScript (ES6), JSON, Responsive Layout, User Experience (UX), Shadcn
Senior Software Engineer
2020 - 2023
Allegis - Cruise Automation - Main
- Led the development of an AV simulator scenario editor app that enabled users to create and visualize scenarios, significantly increasing user engagement.
- Oversaw the WebGL-based renderer optimization, increasing app performance by over two times and improving user experience.
- Created efficient and reusable React components to visualize geospatial data, reducing component development time.
Technologies: React, Node.js, PostgreSQL, JavaScript, React Redux, WebGL, Three.js, OpenGL, Docker, Git, Data Visualization, APIs, Figma, Storybook, Web Development, Functional Programming, TypeScript, HTML5, Back-end Development, Back-end, Back-end Architecture, Full-stack, HTML Canvas, REST APIs, Full-stack Development, Leadership, SVG, CI/CD Pipelines, Agile, Mapbox, REST, Sentry, Google Cloud, Architecture, Scalability, Cloud Architecture, Web Applications, Content Delivery Networks (CDN), Databases, Redis, Dashboards, Web UI, Web UX, Geospatial Data, Minimum Viable Product (MVP), Startups, GitHub, Tailwind CSS, Webpack, API Integration, ChatGPT, Software as a Service (SaaS), Responsive UI, Responsive Web Design (RWD), Zustand, ECMAScript (ES6), JSON, Responsive Layout, Contentful, User Experience (UX), Software Architecture, System Design
Full-stack Software Engineer
2019 - 2020
Vineti
- Integrated Zebra label printers in a RoR back end, resulting in a better logistics experience.
- Added custom PDF generation using templating engines, improving efficiency compared to previous methods and reducing bugs in document generation.
- Created an internal tool to reduce the configuration development time and cut costs for the company.
- Implemented extreme programming methodology within the team, utilizing test-driven development and pair programming to deliver the configurable cell gene therapy platform.
- Developed a SaaS personalized medicine platform using React, Ruby on Rails, and PostgreSQL, resulting in an increase in client engagement and improved patient outcomes.
Technologies: Ruby on Rails (RoR), PostgreSQL, Redux, React, YAML, CSS, Material UI, Test-driven Development (TDD), Extreme Programming, Front-end, Docker, Git, APIs, Figma, Web Development, SCSS, TypeScript, HTML5, Back-end Development, Back-end, Back-end Architecture, Full-stack, REST APIs, Full-stack Development, FastAPI, SVG, WebSockets, CI/CD Pipelines, Agile, REST, Google Cloud, Architecture, Cloud Architecture, Web Applications, Content Delivery Networks (CDN), Databases, Dashboards, Web UI, Web UX, Startups, GitHub, Webpack, API Integration, Software as a Service (SaaS), Responsive UI, Responsive Web Design (RWD), ECMAScript (ES6), JSON, Responsive Layout, User Experience (UX)
Software Developer
2017 - 2019
SoloLearn
- Implemented web code editor and social network postcards, increasing user engagement from browsers.
- Developed a social network for beginner coders, including forums and a Facebook-like feed, increasing online collaboration between users.
- Led successful refactoring of a project to follow best practices of ECMAScript (ES6) and React, resulting in improved maintainability and scalability.
- Built pixel-perfect designs using Material UI and following the best CSS practices.
Technologies: JavaScript, React, Redux, MobX, Material UI, CSS, HTML, Front-end, Git, Data Visualization, D3.js, User Interface (UI), APIs, Web Development, SCSS, HTML5, REST APIs, Agile, REST, Social Media, Web Applications, Dashboards, Web UI, Web UX, Startups, GitHub, API Integration, Responsive UI, Responsive Web Design (RWD), ECMAScript (ES6), JSON, Responsive Layout, User Experience (UX)
Software Developer
2016 - 2017
Renderforest
- Redesigned the landing page using the Pug view engine (previously Jade).
- Tracked and fixed bugs on the online video editor part of the application.
- Developed a thumbnail and preview generators to improve the user experience.
Technologies: JavaScript, Pug, Node.js, React, HTML, CSS, Express.js, Front-end, FFmpeg, Git, User Interface (UI), Web Development, HTML5, Back-end Development, Back-end, Full-stack, Dashboards, Startups, GitHub, Responsive Web Design (RWD), ECMAScript (ES6), JSON, Responsive Layout, User Experience (UX)
Project History
Mechanics Simulations
Small browser-based simulations of physical phenomena such as acceleration and gravitational forces. The project included two sub-projects:
1. Determine if the car can travel past the green street light in time or if the car should decrease the speed. Calculate the acceleration to stop exactly at the traffic light.
2. Simulate planetary motion where the center of the screen has the strongest gravitational pull.
I implemented both of those in a browser using browser animations, React, and HTML5 Canvas.
1. Determine if the car can travel past the green street light in time or if the car should decrease the speed. Calculate the acceleration to stop exactly at the traffic light.
2. Simulate planetary motion where the center of the screen has the strongest gravitational pull.
I implemented both of those in a browser using browser animations, React, and HTML5 Canvas.
Charades App
A React Native mobile game of charades. The project used React Native and read the gyroscope data to decide if the answer is correct (tilt forward) or incorrect (tilt backward). The app timed users and counted the number of right answers at the end. It included several topics for playing and generated random words relating to the topic.
Gradient Descent Simulation
https://github.com/robgev/neural-net-gradient-optimizationImplemented standard and stochastic gradient descent algorithms in the browser to showcase the speed and performance differences of each algorithm. Users could choose the point of the 3D space where the minimum would be. The algorithms would start at some point and try to go to the minimum using the according gradient descent algorithm.
Real Time Chat
A React/Node.js application we built as an experiment for our internal hackerspace usage. The chat had an IBM Watson connection and used WebSockets for communication. The chat also had message editing and terminal-style history storage.
Educational App for Note-taking
https://znotes.org/A web-based app that lets users create and share card-based notes of the classes, collaborate and work on those notes, and create a better way to prepare for exams. It allows using and contributing to the notes generated by other students.
Traffic Flow Optimization Using Quantum Annealers
https://github.com/robgev/quantum-routingThis was my capstone project for my bachelor's degree. I studied and worked with quantum computers such as IBM's Qiskit and DWave's quantum annealers. As my capstone, I wrote a Python app using DWave's quantum annealers to generate an optimized traffic flow for any number of cars given all the roads.
Time Management App
https://github.com/robgev/time-managament-appA React and Node.js app for creating, editing, deleting, and updating tasks, as well as adding tasks for other users as admin, etc. It implements authentication using JWT. The database is PostgreSQL with TypeORM as an ORM service. It also uses Express with middleware and has clean and elegant back-end and front-end solutions.
Pattern Matching Using TypeScript
https://github.com/robgev/pattern-match-tsI worked on implementing functional pattern matching with a syntax similar to OCaml. I implemented everything in TypeScript and documented the development process in an article, available on my blog and Medium. The article got featured on several programming journals in Medium and is featured on Hashnode.
AI Integration for Terminals
https://github.com/robgev/tai-chatA Rust-based terminal UI application that lets you chat with a large language model (LLM). The LLM can be either run locally using ollama or through the network. Implements the Elm architecture for data storage and rendering.
Multithreaded Rust HTTP Server
https://github.com/robgev/rust-http-serverA small multi-threaded HTTP server using Rust to understand the underlying concepts of communicating over TCP.
The server can:
• Work with multiple clients at the same time using threads.
• Save the env context using command line args.
• Handle routes that work with files.
• Be extended by using the abstraction layer to create new routes.
The server can:
• Work with multiple clients at the same time using threads.
• Save the env context using command line args.
• Handle routes that work with files.
• Be extended by using the abstraction layer to create new routes.
Awesome Blog
https://github.com/robgev/simple-blogI created a blog stack from scratch to showcase the capabilities of Supabase and Vite. The blog includes features such as authentication and authorization using JWT tokens, along with clear routing and comprehensive error handling. For the front end, I utilized React/Vite, while the back end is built with Node.js/Express, both using TypeScript. The database is powered by PostgreSQL with Supabase.
Advent of Code
https://github.com/robgev/aoc-2024-rustFor the last two years, I have used Rust to solve Advent of Code problems. Those problems usually feature close to real-life scenarios where you need to come up with an elegant and performant solution in any language. I chose Rust to learn by doing.
VR Apartment Tours
https://bartsnco.com.br/A web application and VR device management service for VR apartment tours.
FEATURES
• Admin panel and editor with Three.js/WebXR for creating new apartment viewings
• Image processing and compression for high-quality VR-ready images
• User/tour management for admins
• Proper authorization and authentication with different roles using Google OAuth and JWTs
• Device management for the root user to manage the fleet of MDM devices
• Remote tour push using AWS
• Full infrastructure architecture in AWS
FEATURES
• Admin panel and editor with Three.js/WebXR for creating new apartment viewings
• Image processing and compression for high-quality VR-ready images
• User/tour management for admins
• Proper authorization and authentication with different roles using Google OAuth and JWTs
• Device management for the root user to manage the fleet of MDM devices
• Remote tour push using AWS
• Full infrastructure architecture in AWS
Full-stack TypeScript Development
http://www.cadstrom.ioI contributed to a CAD tool that helps board prototyping faster through an AI-powered engine. I handled project uploads of different CAD tool types and debugged issues in the automated AI Temporal pipeline.
Education
2016 - 2020
Bachelor's Degree in Computer Science
American University of Armenia - Yerevan, Armenia
Certifications
JULY 2020 - PRESENT
Certificate of Quantum Excellence
IBM
Skills
Libraries/APIs
React, Node.js, REST APIs, Zod, Zod Schema, MobX, D3.js, React Testing Library, Drizzle, Google API, React Redux, WebGL, Three.js, OpenGL, FFmpeg, React Query, OpenAI API, Llama API, llama.cpp, WebXR
Tools
Git, JSS, GitHub, Claude, Slack, Canvas, Figma, Sentry, TanStack Router, Webpack, ChatGPT, Shadcn, OpenSeadragon, Terminal, DocAve, Mongoose, Claude Code, Prisma
Languages
JavaScript, TypeScript, HTML, CSS, Python, HTML5, ECMAScript (ES6), Pug, GraphQL, Sass, SQL, SCSS, Java, YAML, OCaml, Rust, XML
Frameworks
Redux, Material UI, Ruby on Rails (RoR), Express.js, Next.js, Tailwind CSS, Agentic Frameworks, React Native, Jest, Django, LangGraph, JSON Web Tokens (JWT), OAuth 2, Svelte, NestJS
Paradigms
Functional Programming, Agile, REST, Responsive Web Design (RWD), Responsive Layout, Back-end Architecture, Test-driven Development (TDD), Extreme Programming, Role-based Access Control (RBAC)
Platforms
Amazon Web Services (AWS), Vercel, MacOS, Mapbox, Contentful, Docker, Google Cloud Platform (GCP), Firebase, Blockchain
Storage
PostgreSQL, Databases, JSON, MongoDB, Redis, NoSQL, Google Cloud, ArangoDB
Industry Expertise
Social Media
Other
Software Development, Animation, User Interface (UI), Front-end, APIs, Web Development, Full-stack, Full-stack Development, Large Language Models (LLMs), Artificial Intelligence (AI), AI Agents, Architecture, Web Applications, AI Chatbots, Web UI, Web UX, Minimum Viable Product (MVP), Startups, API Integration, Software as a Service (SaaS), Responsive UI, Zustand, User Experience (UX), Solution Architecture, AI Programming, Agentic AI, Software Architecture, System Design, Algorithms, Quantum Computing, Linear Algebra, Physics, Calculus, TypeORM, WebSockets, Data Visualization, Back-end Development, Back-end, HTML Canvas, FastAPI, Leadership, SVG, Supabase, Scalability, Cloud Architecture, Content Delivery Networks (CDN), Retrieval-augmented Generation (RAG), Dashboards, Geospatial Data, OpenAI, Large Language Model Operations (LLMOps), Machine Learning, Mathematics, Qiskit, Storybook, WebAssembly (Wasm), TCP/IP, HTTP, Multithreading, Vite, Supabase Auth, CI/CD Pipelines, Cursor AI, Pattern Matching, Blogs, Anthropic, LangChain, Image Processing, SvelteKit, Optical Character Recognition (OCR), Gemini API, Vibe Coding, Pgvector, ElevenLabs Solutions, SSH
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