// INITIALIZING PIPELINES...

Turning Raw Data into
Actionable Insights

I build the engines that power modern AI. Specializing in high-throughput ETL pipelines, Vector RAG systems, and Autonomous Agent workflows.

↓

Core Competencies

Data Engineering

Designing robust ELT/ETL architectures using Airflow, DBT, and Spark. Ensuring data quality and lineage for downstream AI consumption.

AI Integration

Building RAG (Retrieval-Augmented Generation) systems with Vector Databases. Fine-tuning LLMs for specific domain tasks.

Process Automation

Replacing manual spreadsheet workflows with Python scripts and autonomous agents. If you do it twice, I automate it.

The Engine Room

Tools and technologies I use to bend data.

>> print(stack.current_status)

  • Python
  • SQL
  • Ollama
  • Docker
  • K8s
  • Terraform
  • AWS
  • Azure
  • Airflow
  • Kafka
  • DBT
  • Snowflake
  • TensorFlow
  • Pandas
  • LangChain
  • Pinecone
  • Git
  • n8n

Projects

Things I build: infrastructure, useful tools, and local AI.

Homelab

A Proxmox VE homelab for local AI, data services, and automation, featuring Ollama, Open WebUI, and an NVIDIA GeForce RTX 2070.

Explore the homelab

Yapco

Design personalized planners in Polish or English, customize their pages, and export them for printing in A4 or A5.

Explore Yapco

Audiobook Factory

Turn ebooks into narrated audiobooks using a voice cloned from an audio recording, with optional character voices for dialogue.

Explore Audiobook Factory

Ready to Automate?

Whether you need a custom RAG pipeline or a complete data infrastructure overhaul, let's build systems that scale.

© 2026 Michal Dyzma.