Edwin is available for hire
Hire EdwinEdwin Villacis
Verified Expert in Engineering
Full-stack Developer
Location
Quito, Ecuador
Toptal Member Since
March 29, 2022
Edwin is a full-stack developer with 15 years of experience leading project teams, including in a Fortune 500 company. He specializes in implementing CI/CD pipelines and using diverse programming languages, such as JavaScript, React, Node.js, Java, JEE, Python, HTML-5, CSS, Bootstrap, AWS, RESTful APIs, and React Native. Edwin is also an expert in implementing agile methodologies, namely Scrum, Scrumban, extreme programming (XP), and capability maturity model integration (CMMI).
Portfolio
Freelance
Stable Diffusion, LoRa, Large Language Models (LLMs), Python, Node.js, Django...
Freelance
JavaScript, Java, JEE, Node.js, React, Redux, Bootstrap, Express.js, MySQL...
Telefónica
JavaScript, Java, JEE, Bootstrap, CSS, HTML5, Oracle, Spring, Full-stack...
Experience
Availability
Part-time
Preferred Environment
React, Node.js, Java, Spring, JavaScript, Redux, Spring Boot
The most amazing...
...solutions I've developed are applications for the telecom industry that handle millions of async transactions and serve millions of customers.
Work Experience
Senior Full-stack Developer
2019 - PRESENT
Freelance
- Developed an AI marketplace system for AI-generated images with copyright considerations. Used Stable Diffusion with LoRA parameters for image creation. Made the Python back end with AWS Lambda and ECS.
- Used Next.js, React front end with GraphQL. Integrated Stripe with PayPal.
- Developed AI agents for customer review handling, classification, translation with ChatGPT, parsing, and storage in OpenSearch. Optimized memory usage and enhanced search with vector embeddings. Tested for accuracy. Built back end with Python.
- Created a survey system with JavaScript, React front end, Node.js, Nest.js, and Sequelize back end.
- Hosted on AWS with Lambda for scalability, S3 for storage, and PostgreSQL for secure data storage. Integrated with Qualtrics for survey creation and distribution.
- Made a crawler system with Python, FastAPI, SQLAlchemy, Node.js, Nest.js, Prisma, Metadata in ElasticSearch, back end with ECS, Fargate, and RSS feeds with Lambda. Utilized AI for sentiment analysis, topic modeling, and GPT-based summarization.
- Created a Crawler back end with Node.js, deployed on AWS Lambda for scalable operations. Used PostgreSQL via Sequelize for database interactions. Crafted the front end with TypeScript, React, and Redux for state management. Styled with Bootstrap.
- Used OCR tech to convert legal PDFs to text, processed with Chat GPT for NER data stored in Elasticsearch. Users filter and visualize results in a knowledge graph, annotated documents by PDFTron. Used AWS Lambda for the back end and GraphQL on ECS.
- Utilized React and TypeScript for the front end and Node.js, MongoDB, and Express. Used GCP for security and scalability for the back end. Utilized React Native for the mobile and Material UI.
- Developed security, distribution, shipment, and logistics. Designed the app for order management and warehouse optimization.
Technologies: Stable Diffusion, LoRa, Large Language Models (LLMs), Python, Node.js, Django, Machine Learning, FastAPI, React, TypeScript, JavaScript, PyTorch, LangChain, Jupyter Notebook, Google Colaboratory (Colab), RunPod, Amazon Web Services (AWS), Amazon RDS, PostgreSQL, Lambda Functions, Amazon Cognito, Amazon Elastic Container Service (Amazon ECS), Elasticsearch, GraphQL, Apollo, PDFTron, Flask, AWS Step Functions, Amazon Simple Queue Service (SQS), REST APIs, Google Cloud Platform (GCP), AWS Amplify, Amazon S3 (AWS S3), Amazon Aurora, MongoDB, Next.js, Tailwind CSS, Figma, Amazon Titan, Bedrock, Stripe API, PayPal API, ChatGPT, NestJS, Sequelize, AIM, Qualtrics, SQLAlchemy, Prisma, AWS Fargate, Redux, React Query, Bootstrap, React Native, Expo, Material UI, Express.js, Styled-components, Jest, Sass, Prompt Engineering, Startups, Artificial Intelligence (AI), Natural Language Processing (NLP), Search Engines, Elastic, Chatbots, OpenAI, AI Integration, JSON
Senior Full-stack Developer
2014 - 2019
Freelance
- Developed, maintained, and coded a React and Bootstrap application that managed tokens identified via OCR technologies from over 800,000 billing images hosted on AWS ECS.
- Created an application to control and push surveys offered to mobile users, track the responses, handle phone and internet bills, parse the OCR data, do surveys for eCommerce sites, and maintain the AWS infrastructure.
- Maintained a React Native app collecting minute-by-minute telemetry from over 100,000 users, saved in SQLite, and sent to a terabyte Redshift database. Supported data science analysis using Python for US Telecom decision-making.
- Coded and adapted a Telemetry React Native application for a desktop environment using Electron, React Native for Web, and React.
- Made a serverless microservices app handling over 200,000 surveys per month and OCR-processed bills. Utilized AWS Lambda in Java, Spring, and Hibernate with 203+ Lambda functions coordinated through AWS Step Functions.
- Created and maintained a back-end system using NiFi and Java, aggregating more than two terabytes of telemetry data in a large-scale Redshift database and storing the aggregated outcomes in PostgreSQL.
- Made control software for chemical production, managing oxygen stabilization, production tracking, and raw materials. Used React with Bootstrap, Python, and MongoDB on AWS EC2. Implemented AWS changes tailored to GraphQL APIs for managerial queries.
- Architected an app for electrical meter control and monitoring, analyzing consumption and efficiency, utilizing React, Node.js, GraphQL, and MongoDB. Optimized with Redux, Apollo, and Docker. Implemented Jest, Cypress. Deployed on AWS EC2.
- Developed the accounting and payroll system with TypeScript, React, and Redux for UI. Utilized Node.js for server logic and MySQL for data storage. Managed Linux server deployment. Enhanced financial management efficiency ensured secure transactions.
- Made electrical project control software with React, Bootstrap front end, Node.js, and PostgreSQL back end. Employed Jest for testing. Managed Linux server deployment. Enhanced project management, stock control, payroll, and accounting efficiency.
Technologies: JavaScript, Java, JEE, Node.js, React, Redux, Bootstrap, Express.js, MySQL, MongoDB, Git, HTML5, Spring, Spring Boot, React Redux, Jest, Selenium, JUnit, Full-stack, TypeScript, REST APIs, Microservices Architecture, Responsive Layout, Cross-browser Compatibility, Full-stack Development, PostgreSQL, Front-end, Amazon Web Services (AWS), Docker, HTML, APIs, MariaDB, Event-driven Programming, User Interface (UI), GitHub, UI Components, Integration Testing, Serverless Architecture, Next.js, NestJS, GraphQL, API Integration, Integration, Security, AWS CLI, Amazon Simple Queue Service (SQS), Amazon Simple Notification Service (Amazon SNS), AWS Lambda, Amazon EC2, Electron, Amazon CloudWatch, Amazon RDS, Lambda Functions, CI/CD Pipelines, Prisma, Tailwind CSS, Async Batch Processes, Payment APIs, Amazon API, AWS CloudFormation, Sequelize, Redshift, Amazon S3 (AWS S3), OpenAI GPT-3 API, OpenAI GPT-4 API, Back-end, NoSQL, REST, Cloud, Google Cloud Platform (GCP), Microservices, JSON Web Tokens (JWT), Scalability, API Gateways, ETL, Message Queues, Authentication, Middleware, AWS SDK, Python, React Native, Firebase, Mobile, AWS AppSync, Apollo, Elasticsearch, PDFTron, Amazon Cognito, Auth0, Apache NiFi, NiFi, Angular, AngularJS, Jira, Technical Architecture, Scraping, Web Scraping, Swagger, Amazon Elastic Container Service (Amazon ECS), Containerization, Multitenancy, AWS Amplify, Front-end Development, Figma, PWA, Webpack, Payment Gateways, Stripe, Stripe Checkout, Algorithms, Minimum Viable Product (MVP), Team Leadership, SQLite, Artificial Intelligence (AI), Data Science, JSON, XML
Software Development Manager
2006 - 2014
Telefónica
- Led 135 employees within a development and testing team in a culture of continuous improvement.
- Received recognition within the top two software delivering processes in LATAM.
- Increased software development quality from 83% to 97%.
- Reduced the time-to-market by 74% from user requirement definition to deployment and acceptance in production.
Technologies: JavaScript, Java, JEE, Bootstrap, CSS, HTML5, Oracle, Spring, Full-stack, TypeScript, Full-stack Development, Front-end, WebSockets, HTML, Event-driven Programming, User Interface (UI), GitHub, Integration Testing, API Integration, Integration, APIs, Security, Architecture, Async Batch Processes, Back-end, API Gateways, ETL, Message Queues, Authentication, Middleware, Technical Leadership, Requirements Analysis, Project Portfolio Management (PPM), CTO, Technical Architecture, Containerization, Front-end Development, Webpack, jQuery, Algorithms, Team Leadership, Oracle Database, JSON, XML
Senior Full-stack Developer
1999 - 2006
Colvida SA Life Insurance Company
- Implemented the first banking insurance system in the country.
- Carried out the first educational insurance system in the country.
- Increased sales up to 25% yearly in individual products.
Technologies: JavaScript, Oracle, SQL, PL/SQL, Linux, Windows Server, Full-stack, Full-stack Development, Front-end, HTML, User Interface (UI), Architecture, DNS, ETL, Technical Leadership, Requirements Analysis, Technical Architecture, Algorithms, Team Leadership, Oracle Database
Full-stack Developer
1996 - 1999
Integral SA Insurance Company
- Developed the company's insurance core system from scratch.
- Created the company's business intelligence system for decision making.
- Decreased the processing time up to 50% in regulatory information.
Technologies: JavaScript, CSS, HTML5, SQL, PL/SQL, Oracle, Linux, Full-stack Development, Front-end, HTML, DNS, ETL, Requirements Analysis, Algorithms, Team Leadership, Oracle Database
Experience
AI-powered Image Marketplace
I architected and developed an advanced AI marketplace system to generate and manage AI-generated images with copyright awareness, utilizing stable diffusion for image creation. The back end, built on Python, uses AWS Lambda for scalable processing, with AWS Step Functions and SQS for workflow orchestration. To support resource-intensive AI tasks, Amazon ECS and EC2 were seamlessly integrated, while AWS API Gateway secured system access. User authentication is ensured via Google Cloud Identity and Amazon Cognito, with AWS Amplify and S3 managing front-end development and asset storage. Data is stored in Amazon RDS Aurora PostgreSQL and MongoDB, with FastAPI and Django for backend APIs.
The front end utilizes Next.js, React, and Figma for a user-friendly interface enhanced by Tailwind CSS. For advanced image searches, I implemented AWS OpenSearch, Titan, and Bedrock. GraphQL facilitates data interactions, and Stripe and PayPal manage financial transactions, offering a comprehensive solution for image marketplace needs.
The front end utilizes Next.js, React, and Figma for a user-friendly interface enhanced by Tailwind CSS. For advanced image searches, I implemented AWS OpenSearch, Titan, and Bedrock. GraphQL facilitates data interactions, and Stripe and PayPal manage financial transactions, offering a comprehensive solution for image marketplace needs.
Enhancing Customer Review Analysis System with AI-powered Tools
I created AI agents to revamp our approach to customer reviews, focusing on classification, translation, parsing, optimization, search capabilities, evaluation, and the underlying technology stack. The agents classify reviews for precise analysis and, with ChatGPT's LLM, translate content, breaking language barriers. Original reviews are stored in DynamoDB, and processed reviews are stored in OpenSearch, enabling efficient data retrieval. We optimized memory usage and token count for cost efficiency. Our customer service leverages vector embeddings for accurate searches in OpenSearch. Rigorous testing and AI agent chaining ensured high performance. The system, powered by Python, LangChain, AWS Lambda, API Gateways, and Amazon Titan, enhances feedback analysis, aiding in smarter decision-making and boosting customer satisfaction.
Web Crawler System
I developed a crawler system using a back-end built in Python and Node.js that gathers over a decade's worth of data from news sources, blogs, social media, and RSS sources. The back-end system is designed to run within Docker containers, which are orchestrated by AWS ECS, utilizing AWS Fargate to execute these containers, which allows a serverless operational model. The RSS feeds are aggregated using a back-end system developed as AWS Lambdas. These articles are sent to an AWS SQS, then processed as needed, and subsequently stored in an AWS S3 bucket in its metadata in an Elasticsearch database. The system handles a significant volume of data, making over 100,000 requests to various sources. Each piece of retrieved content undergoes processing through a sophisticated AI-based framework. This includes a sentiment analysis model, which evaluates the tone and sentiment of the text. An AI-driven topic modeling component is employed to categorize and understand the primary themes and subjects of the content. This analysis empowers investors with a nuanced understanding of the current market sentiment and key trends. Furthermore, the back-end system utilizes a GPT model to summarize the articles, storing them in Elasticsearch.
Web Crawler Administration System
I designed the architecture of and constructed a Crawler Administration System to oversee and manage multiple web crawlers operating across various Docker containers. This system is adept at gathering and presenting real-time statistics about active crawlers and flags that encounter errors. The back end was crafted using Node.js lambda functions integrated with an RDS PostgreSQL database. For database interaction and object-relational mapping, Sequelize is employed. The front-end interface, designed for monitoring the crawler machines, was built using TypeScript, React, Redux, React Query, and Bootstrap. This interface provides administrators with a clear and interactive dashboard to monitor the crawler machines. The front end communicates with the back end via RESTful API gateways.
RSS Feed Aggregator
I developed an RSS aggregator using Python and Node.js, Lambda functions, SQLAlchemy, and Amazon RDS PostgreSQL. This aggregator collects RSS-focused news and sends it to an AWS SQS queue for ingestion into the AI system. The front-end control and maintenance interface was developed using TypeScript, React, and Bootstrap.
PDF Viewer and Annotator
I architected and implemented a comprehensive system that transforms legal litigation PDF documents—often exceeding 100 pages—into text through Optical Character Recognition (OCR). This text is then processed using an AI framework, specifically ChatGPT, to perform Named Entity Recognition, efficiently extracting key entities and information. The AI processing leads to the creation of concise summaries of these extensive documents.
This system features a user-friendly interface and is developed using React, TypeScript, and Material-UI, which allows users to manually annotate entities within the PDFs using PDFTron. These annotations, along with AI-processed data, are stored in an Elasticsearch database, ensuring efficient data management and retrieval. The processed information is made accessible to the front end through a Node.js application, utilizing GraphQL and Apollo, all hosted within an AWS ECS service for robust scalability and performance.
To secure the system, I integrated AWS Cognito for robust user authentication and authorization. This security layer ensures that only authorized users can access and interact with the system, maintaining the confidentiality and integrity of the legal documents.
This system features a user-friendly interface and is developed using React, TypeScript, and Material-UI, which allows users to manually annotate entities within the PDFs using PDFTron. These annotations, along with AI-processed data, are stored in an Elasticsearch database, ensuring efficient data management and retrieval. The processed information is made accessible to the front end through a Node.js application, utilizing GraphQL and Apollo, all hosted within an AWS ECS service for robust scalability and performance.
To secure the system, I integrated AWS Cognito for robust user authentication and authorization. This security layer ensures that only authorized users can access and interact with the system, maintaining the confidentiality and integrity of the legal documents.
Automated Token Identification and Management System
I handled the development, maintenance, and coding of the front end for an application designed to identify and manage tokens in PDFs and image-based bills. This application, which processes over 800,000 billing documents, is built using React, JavaScript, TypeScript, Redux, and Bootstrap and is hosted on AWS EC2. The identified tokens are stored in a PostgreSQL database, supported by a back end developed in Java with Spring Boot. This back end leverages AWS Lambda functions, API gateways, and REST APIs for efficient operation. Additionally, all images and PDFs are securely stored in AWS S3. The application's security is robustly managed through AWS Cognito and IAM, ensuring protected access and data integrity.
Scalable Survey System
Developed a dynamic survey system to effectively engage our customer base. The front end is crafted with JavaScript and React for a seamless user experience, while the back end runs on Node.js, utilizing Sequelize as the ORM for smooth PostgreSQL interactions, ensuring efficient survey data management.
Hosted on AWS, the system employs Lambda for scalable serverless operations, API Gateway for secure API management, and S3 for reliable storage, including the front-end application. AWS IAM meticulously controls access, enhancing security. PostgreSQL serves as a secure database, enabling detailed data analysis.
This setup guarantees user-friendly survey participation, scalable processing, and strong data security, streamlining survey creation and distribution.
Hosted on AWS, the system employs Lambda for scalable serverless operations, API Gateway for secure API management, and S3 for reliable storage, including the front-end application. AWS IAM meticulously controls access, enhancing security. PostgreSQL serves as a secure database, enabling detailed data analysis.
This setup guarantees user-friendly survey participation, scalable processing, and strong data security, streamlining survey creation and distribution.
Mobile Phone Call Control
It is a time-window control for calls in the prepaid telephone system, based in Java. The application performed with great accuracy, controlling the service provisioning in the core telephony central.
Middleware to Reload the Cellphone Mobile Wallet
It is a Node.js and Express middleware application that uses a mobile wallet to allow the reloading of mobile customer phones in remote locations such as retail stores, pharmacies, supermarkets, and grocery stores. The process works with async transactions for providing interfacing between the prepaid telephone system and the user’s mobile phone.
Skills
Languages
JavaScript, HTML5, CSS, SQL, TypeScript, HTML, Python, XML, GraphQL, Java, Sass
Frameworks
Redux, Bootstrap, Express.js, JSON Web Tokens (JWT), Spring Boot, Jest, Selenium, JUnit, Next.js, NestJS, Spring, Tailwind CSS, React Native, Angular, AngularJS, Bedrock, Django, Electron, Swagger, Material UI, Flask
Libraries/APIs
React, Node.js, jQuery, React Redux, REST APIs, Amazon API, Stripe, AWS Amplify, Stripe API, PayPal API, SQLAlchemy, PyTorch
Tools
Git, GitHub, AWS CLI, Amazon Simple Queue Service (SQS), Amazon Simple Notification Service (Amazon SNS), Amazon CloudWatch, AWS CloudFormation, Sequelize, AWS SDK, Amazon Cognito, Auth0, Jira, Amazon Elastic Container Service (Amazon ECS), Webpack, AWS IAM, AWS Step Functions, MongoDB Atlas, Stripe Checkout, Elastic, Prisma, AWS AppSync, PDFTron, Apache NiFi, Figma, AWS CodeBuild, AWS CodeDeploy, Amazon OpenSearch, AWS Fargate, Qualtrics, Expo
Paradigms
Microservices Architecture, Event-driven Programming, Serverless Architecture, REST, Microservices, ETL, Requirements Analysis, Lambda Architecture, Data Science, Responsive Layout, Compiler Design
Platforms
Oracle, JEE, Amazon Web Services (AWS), Docker, AWS Lambda, Amazon EC2, Oracle Database, Linux, Windows Server, Google Cloud Platform (GCP), Mobile, Firebase, Jupyter Notebook
Storage
MongoDB, PL/SQL, PostgreSQL, MariaDB, Amazon S3 (AWS S3), NoSQL, Amazon Aurora, SQLite, JSON, MySQL, Redshift, Elasticsearch, Amazon DynamoDB
Other
OOP Designs, Software Architecture, People Management, Operations, Full-stack, Full-stack Development, Front-end, APIs, User Interface (UI), Integration Testing, API Integration, Integration, Amazon RDS, Lambda Functions, Architecture, OpenAI GPT-3 API, OpenAI GPT-4 API, Back-end, Cloud, API Gateways, Message Queues, Authentication, Middleware, Technical Leadership, CTO, Technical Architecture, Containerization, Front-end Development, Amazon API Gateway, Relational Database Services (RDS), ECS, AWS CodePipeline, Payment Gateways, Algorithms, Minimum Viable Product (MVP), Team Leadership, Startups, Natural Language Processing (NLP), Search, Search Engines, natural language search, Full-text Search, Chatbots, OpenAI, AI Integration, Artificial Intelligence (AI), Statistics, Cross-browser Compatibility, WebSockets, UI Components, Security, CI/CD Pipelines, DNS, Async Batch Processes, Payment APIs, Scalability, Apollo, NiFi, Project Portfolio Management (PPM), Multitenancy, PWA, Springbot, OCR, Images, Identity & Access Management (IAM), Large Language Models (LLMs), ChatGPT, LangChain, Amazon Titan, FastAPI, Stable Diffusion, LoRa, Embeddings from Language Models (ELMo), Machine Learning, Prompt Engineering, Scraping, Web Scraping, RSS Feeds, Apollo Server, PDF, Google Colaboratory (Colab), RunPod, AIM, React Query, Styled-components
Education
2006 - 2007
Master's Degree in Business Administration
Catholic University of Chile - Santiago, Chile
1988 - 1994
Bachelor's Degree in Computer Science
National Polytechnic School - Quito, Ecuador
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