Glib S.

Full-stack AI engineer for production SaaS: tool-using AI agents, secure RAG, and document automation built with React, Node, and Python.

$167K+ earned | 100% Job Success | Top Rated Plus | 2,400+ Upwork hours | 8+ years of engineering experience

I help founders and product teams turn an AI idea, fragile prototype, or AI-generated codebase into a dependable product people can use. I own the work end to end: product workflow, UI, APIs, data model, AI integration, permissions, retrieval quality, deployment, cost, and reliability.

WHAT I BUILD
• AI agents that use real tools, APIs, and internal data, not chat demos.
• Secure RAG and knowledge systems with citations, access control, evaluation, and practical fallback behaviour.
• Document-AI workflows that turn PDF, Word, Excel, and messy operational data into structured, reviewable outputs.
• Full-stack AI SaaS products: web app, backend, authentication, database, billing, admin tools, and deployment.
• Rescue and modernization work for existing or AI-generated codebases that need to work beyond the demo.

SELECTED PRODUCTION WORK
• Built a partner-discovery SaaS with a Claude tool-use agent that researches and qualifies 10,000+ companies, then returns cited answers.
• Built an enterprise RAG system over internal documents, with group-based access control enforced at the PostgreSQL/pgvector layer.
• Built a document-AI platform that turns supplier PDF, Word, and Excel specifications into structured product data and completed retailer forms.
• Built AI-powered web and mobile products, including a caregiver copilot with structured summaries and safeguards for sensitive data.
• Built computer-vision systems that improved detection accuracy by 36% and reduced manual QA by 65%.

HOW I WORK
I start with the business outcome and the real constraints: source data, user roles, reliability, security, latency, cost, and how the feature will be operated after launch. You get working increments, direct communication, clear tradeoffs, and maintainable code—not a black-box prototype.

CORE STACK
React, Next.js, TypeScript, Node.js, NestJS, Python, FastAPI, PostgreSQL, MongoDB, AWS, Docker, REST and GraphQL APIs.

For AI: OpenAI, Claude, Gemini, LangChain, embeddings, vector search, RAG, tool calling, structured outputs, and applied ML/computer vision when the product needs it.

My background includes R&D at SoftServe and a PhD in Artificial Intelligence Systems in progress. I bring research discipline where it helps, but focus on shipping useful software.