I build custom LLM pipelines and AI agents using Python and modern language models - clean, production-ready code, not just a demo.
From a simple RAG system that answers questions from your documents, to a multi-step agent that automates complex workflows, to a full AI system with tool use, memory, and API integrations.
What I deliver:
- Clean Python code with detailed comments and documentation
- RAG (document Q&A), multi-agent pipelines, tool-calling systems
- Works with GPT-5.5, Claude Sonnet 4.6 / Opus 4.8, Gemini 3.5 Flash, DeepSeek v4
- LangChain / LangGraph, OpenAI API, Anthropic API, custom integrations
- Source code + setup instructions included
Background: PhD in Computer Engineering (NLP/LLM focus, Taras Shevchenko National University of Kyiv). Research specialization: LLM fine-tuning and coreference resolution.
Typical use cases:
- Document Q&A chatbot (RAG over PDFs, databases, web pages)
- AI agent with tool use (web search, code execution, API calls)
- Automated report generation or data extraction pipeline
- Multi-step workflow with memory and context management
- Telegram / Slack bot backed by an LLM
Please message before ordering to confirm your use case and scope.