🧠
Applied AI engineering
Designing AI solutions around LLMs, prompting, retrieval, agentic workflows, quality evaluation and connections to real data sources.
Applied AI · RAG · Agentic systems · Python
I build AI systems, not just nice prompts.
I am a Python developer with a strong focus on Applied AI. I currently work heavily with RAG systems, vector databases, agentic platforms, local AI solutions, AI evaluation, LangChain, LangFuse, relational databases, Docker and server infrastructure around real AI applications.
About me
A developer who connects AI, data and infrastructure.
I have experience in application development, backend services, database solutions, data pipelines and automation. Today I focus mainly on practical AI implementation: from data ingestion, processing and indexing, to response evaluation and server-side operation.
🗄️
Data, databases and retrieval
Working with relational databases, vector databases, embeddings, document chunking, data flows and retrieval of relevant context for AI systems.
⚙️
Server-side solutions
Docker, Linux/VPS, API services, background jobs, rate limiting, logging, monitoring, integrations and backend infrastructure for stable technical systems.
AI focus
I focus on AI topics that are closer to real production systems.
Not only “which model is the best”, but the whole system around the model: data, retrieval, tools, evaluation, secure integrations, infrastructure and feedback loops.
🔎
RAG systems
Retrieval-Augmented Generation, document bases, chunking, embeddings, hybrid search, context management and reducing hallucinations through better data.
🧬
Vector databases
Knowledge indexing, semantic search, embeddings, metadata filtering and designing the data layer for searching internal documents or a company knowledge base.
🤖
Agentic systems
AI agents, tool calling, MCP concepts and connecting models to Slack, Notion, APIs, databases, file systems or custom backend tools.
📏
AI evaluation
Testing answer quality, tracking regressions, comparing models, datasets and prompts, and using observability tools such as LangFuse.
🧰
LangChain & orchestration
Composing AI workflows, chains, agents, retrievers, tools and data flows into functional applications that can be extended further.
💻
Local AI solutions
Exploring local LLMs, model deployment on own hardware, working with VRAM limits, privacy-first solutions and local inference integrations.
Technical stack
A stack that connects AI applications with the real world.
I am most interested in the place where AI, Python backend, data layers, infrastructure and practical automation meet.
AI / LLM / RAG
Backend / data / infra
Focus & projects
The areas I currently enjoy the most.
I prefer solutions where AI is not an isolated chatbot, but a part of a system: with data, tools, rules, evaluation, logging and a clear deployment path.
Company knowledge bases and RAG over documents
Designing a pipeline from documents through chunking, embeddings, vector index, metadata filters, retrieval strategy and model responses with quality checks and sources.
Agentic platforms and AI workflows
Connecting LLMs with tools, APIs, databases and internal systems. I am especially interested in turning an AI agent into a practical helper, not a random experiment.
Evaluation and observability of AI applications
Tracking prompts, answers, costs, latency, retrieval quality and regressions between versions. Without evaluation, an AI system is only subjectively good.
Local and privacy-first AI solutions
Exploring locally running LLMs, their limits, deployment on own hardware and practical use cases where data should not leave owned infrastructure.
Data pipelines and financial datasets
Processing large datasets, SEC EDGAR, 13F/NPORT data, CUSIP mapping, PostgreSQL, time series, portfolio analytics and visualizations.
Server services, Docker and backend infrastructure
Linux/VPS setups, Docker, API services, async request handlers, rate limiting, logging, cron jobs and custom technical services for data and AI workflows.
Education & certifications
My AI knowledge did not appear out of nowhere. It is backed by hours of study, programming and practical assignments.
This is a selection of courses, specializations and certificates that form my AI/ML foundation: machine learning, NLP, generative AI for software development, agentic systems and knowledge graphs for RAG.
Complete specialization
Machine Learning Specialization
A core foundation for supervised learning, regression, classification, neural networks, recommender systems, unsupervised learning and reinforcement learning.
Supervised Machine Learning: Regression and Classification
Unsupervised Learning, Recommenders, Reinforcement Learning
Advanced Learning Algorithms
Natural Language Processing
NLP with Classification and Vector Spaces
Text classification, vector representations, working with text and NLP fundamentals that are also important for embeddings, semantic search and RAG systems.
Natural Language Processing
NLP with Probabilistic Models
Probabilistic models in NLP, language modeling and additional techniques needed for deeper understanding of natural language processing.
Complete specialization
Generative AI for Software Development
A specialization focused on practical use of generative AI in software development, team collaboration, system design and modern AI-assisted engineering.
Introduction to Generative AI for Software Development
Team Software Engineering with AI
AI-Powered Software and System Design
Agentic AI
Agentic AI
A course focused on agentic AI systems, their design, tool use, workflows and how AI models can be connected to practical multi-step tasks.
RAG & Knowledge Graphs
Knowledge Graphs for RAG with Neo4j
A course on using knowledge graphs in RAG systems, connecting structured knowledge, graph databases and retrieval layers for more accurate AI answers.
I treat these certificates as evidence of systematic study, not just badges. They represent hours of work, implementation, experimentation and practical programming in areas I now use when designing AI solutions.
Experience
The foundation is still software development and data.
Current
Applied AI / RAG / Agentic systems
Working on RAG systems, agentic workflows, vector databases, local AI models, AI evaluation, the LangChain/LangFuse ecosystem and server infrastructure around AI applications.
2021 – 2022
Data Scientist / Developer
Development of ML solutions and data tools for financial markets, time-series processing, portfolio UI and automated workflows.
2019 – 2022
Freelance programmer
Automation of trading strategies, development of tools for MT4/MT5, Python integrations, Binance API, Telegram, Discord, TradingView and system integrations via REST/TCP/HTTP.
Long-term
Python / C++ / backend / databases
Development of applications, server services, internal tools, database solutions, data scripts, automation and technical integrations.
Contact
Let’s build a practical AI solution.
I am interested in projects around RAG systems, AI agents, local LLMs, AI evaluation, vector databases, backend services, data pipelines and automation.
info@peterbaksa.sk