Mateusz P.

Overview I’m an AI engineer and full‑stack developer with 8+ years of experience delivering production software. My focus is on AI agents, retrieval‑augmented generation (RAG) pipelines and large language model (LLM) systems that transform unstructured data into actionable insights and automate multi‑step work. I design, build and deploy robust solutions that power AI agents, intelligent search, conversational assistants and decision‑support tools. Why work with me? I combine deep technical expertise with a pragmatic approach to delivering business value. I’ve helped clients implement AI agents, RAG‑powered knowledge bases, custom chatbots and LLM‑orchestration frameworks that significantly improve search accuracy and user satisfaction. Clients trust me for clear communication, on‑time delivery and high‑quality results. What I do best AI agents: Build LLM‑powered agents that plan, use tools and APIs, and complete multi‑step tasks, with LangGraph, LangChain and function calling. I add memory, guardrails, human‑in‑the‑loop steps and evaluation so agents are reliable in production. RAG & LLM systems: Design and implement end‑to‑end RAG pipelines, from document ingestion and embedding generation to vector search and LLM inference. Experienced with LangChain, LlamaIndex, LangGraph, GPT‑4, Llama 3 and open‑source models. Vector databases & data engineering: Build scalable semantic‑search infrastructures using pgvector, Pinecone, Weaviate and Elasticsearch, handling millions of documents across diverse formats. ML & MLOps: Deploy models and services with FastAPI, Docker, and CI/CD; set up monitoring, logging and evaluation frameworks to ensure reliability in production. Full‑stack AI applications: Develop modern front‑ends using React/TypeScript and integrate AI features and agents seamlessly into web platforms. Proven results Developed a RAG‑powered search tool for a legal firm, which ingested thousands of case documents and delivered relevant passages via a GPT‑4 reasoning layer—dramatically reducing researchers’ time to find answers. Built an internal knowledge‑base chatbot using LangChain, LlamaIndex and Pinecone to serve engineers at a tech startup; the bot achieved high accuracy and increased productivity by enabling instant access to technical documentation. Led the migration of multiple LLM projects to production, establishing robust MLOps pipelines, reducing inference latency and improving system stability. Tools & technologies Python, FastAPI, LangChain, LlamaIndex, LangGraph, GPT‑4 / OpenAI API (function calling), Llama 3, Hugging Face Transformers, Pinecone, Weaviate, pgvector/PostgreSQL, Elasticsearch, Docker, Kubernetes, Celery/RQ, GitHub Actions, AWS/GCP. Client success & reviews I consistently receive 5‑star feedback for my technical excellence, clarity of communication and ability to solve complex problems. Past clients have praised my commitment to delivering robust solutions and my willingness to go the extra mile to ensure project success. Let’s build smarter AI together If you need someone who can design, implement and deploy an AI agent or a custom RAG pipeline, fine‑tune an LLM or integrate AI into your product, let’s talk. You can book a consultation to discuss your goals in detail or send me a message to start your project.