vix95
/typenova
A local-first desktop Markdown editor where every heading becomes a node of an explorable graph.
AI Engineering Notebook
Hi, I'm Paweł.
I build applied AI systems.
I experiment with LLMs, machine learning, deep learning, and data science. This is my portfolio, blog, and technical journal of things I learn and build.
Areas I'm exploring
Language systems
LLM
I work with language models, prompt engineering, and response evaluation.
Knowledge retrieval
RAG
I build retrieval-based systems and knowledge bases: vector search, hybrid retrieval, sourced answers.
Agentic systems
Autonomy
Tool-using agents, planning loops, memory, evaluation, and controlled automation with LLM-based systems.
Patterns & validation
Machine Learning
Classical models, feature selection, validation, and interpretation of results.
Neural architectures
Deep Learning
Neural networks, CNNs, RNNs, transformers, and representation learning.
Data exploration
Data Science
Data analysis, visualization, statistics, and dataset exploration.
Visual intelligence
Computer Vision & OCR
Image understanding, document parsing, text extraction, layout analysis, and vision-language pipelines.
Generative media
Image & video models
Diffusion, multimodal generation, and model workflows for visual content understanding and creation.
Systems thinking
Modelling
Solution design, ML pipelines, and experiment prototyping.
Personal projects
View all projects
MobiSkan.pl
The public product site for MobiSkan, built on Joomla with a fully custom template. The only parts of Joomla core actually reused are the CMS content…
View project
MobiSkan WMS
MobiSkan WMS is the server-side counterpart to the MobiSkan Android application. It turns individual CipherLab data collectors into one coordinated wa…
View project
MobiSkan 3.0
MobiSkan is an offline-first Android application for warehouse staff carrying CipherLab data collectors (RK25, RK26, RK95): receiving goods, issuing s…
View projectFeatured open source
View all open source
Latest posts
View all posts- Building a Production Website Chatbot with RAG: An Advanced Engineering Guide →
- REST and WebSocket in AI Systems: Designing Request-Response and Real-Time Communication →
- Random Forest From First Principles: Bagging, OOB Evaluation, Tuning, and Interpretation →
- A Churn Model Does Not Tell You Whom to Target: Uplift Modeling with a T-Learner →
- Embedding Drift in Retrieval Systems: Why Centroid Shift Is Not Enough →
- Causal Overlap Before Estimation: Diagnosing Support, Trimming, and Estimand Change →
My journey
View all- Apr 2026 - Present AI ENGINEER Sea Machines Robotics
- Dec 2025 - Apr 2026 SENIOR DATA SCIENTIST Grid Dynamics
- Jul 2022 - Dec 2025 DATA SCIENTIST Flex
- 2022 COMPUTER SCIENCE MASTER'S DEGREE University of Gdańsk
- 2020 COMPUTER SCIENCE BACHELOR’S DEGREE University of Gdańsk
My latest courses
Educative · Employer-Sponsored · 2026
Focused on architecting scalable, production-grade ML systems. Covered end-to-end pipeline design, infrastructure trade-offs, and reliability considerations. Emphasized latency, monitoring, and deployment best practices derived from large-scale industry environments.
Udemy · Independent Study · 2026
Designed multi-agent architectures using LangGraph, CrewAI, and AutoGen. Implemented tool orchestration, task routing, and Docker-based deployments. Delivered complex agent collaboration systems including autonomous trading simulations.
LinkedIn · Employer-Sponsored · 2026
Explored transformer architectures and post-GPT-3 LLM advancements. Analyzed architectural evolution and real-world NLP applications. Strengthened practical understanding of LLM deployment strategies.
Udemy · Independent Study · 2026
Built responsive web applications using semantic HTML5 and modern CSS (Flexbox, Grid). Applied structured layout systems and UI/UX principles. Strengthened frontend debugging and optimization skills.
Udemy · Independent Study · 2025
Implemented distributed MapReduce jobs in Python (MRJob) within the Hadoop ecosystem. Executed large-scale workloads locally and on AWS EMR. Applied NLP and analytical processing to real-world Big Data datasets.
Contact
Collaboration inquiry
Let’s discuss AI, ML, or technical collaboration.
I am open to AI engineering roles, applied machine learning projects, technical collaboration, and focused project discussions.
pawel.labuda@itvix.plSkills & expertise
Core AI & Machine Learning
Vector Search, Databases & Data Platforms
- Snowflake
- Oracle
- Microsoft SQL Server
- PostgreSQL
- MySQL
- SQLite
- Teradata
- MongoDB
- Firebase
- OpenSearch
- Elasticsearch
- Weaviate
- Chroma
PyData & Scientific Computing
- NumPy
- Pandas
- Polars
- Dask
- SciPy
- Matplotlib
- Seaborn
- Statsmodels
- Pydantic
Machine Learning, Deep Learning & NLP
- Scikit-learn
- XGBoost
- LightGBM
- CatBoost
- PyTorch
- TensorFlow
- Keras
- Hugging Face Transformers
- Sentence-Transformers
- NLTK
- BERTopic
- A/B testing
- LoRA/QLoRA/PEFT fine-tuning
AI Agents & LLM Orchestration
- OpenAI API
- Ollama
- LangChain
- LangGraph
- n8n
Engineering & Production Systems
MLOps, DevOps & Automation
- Docker
- Git
- GitHub Actions
- GitLab CI/CD
- Gitea
- Azure DevOps
- MLflow
- Poetry
- uv
- Conda
- pip
Backend & API Engineering
- FastAPI
- Flask
- Django
- REST
- GraphQL
- WebSockets
- JWT
- JSON
Development Tools & IDEs
- Jupyter Notebook
- Google Colab
- Android Studio
- PyCharm
- GoLand
- VS Code
- Postman
- Minitab
Software Engineering Principles
- Concurrent & Asynchronous Systems
- Object-Oriented Design (OOP)
- Design Patterns
- Test-Driven Development (TDD)
- RESTful Service Design
- Microservices
- Distributed Architecture
Applied Data & Product Context
Programming
- Python
- Java for Android
- Go
- C++
- SQL
Data Visualization & BI
- Tableau Desktop
- Tableau Prep
- Alteryx Designer
- Prometheus
- Grafana
- Dash
- Streamlit
- PySpark
Collaboration & Delivery
- Slack
- Jira
- Confluence
- Nextcloud
- Scrum
- Kanban
- Microsoft Teams
Frontend & Interactive Interfaces
- HTML5
- CSS3
- Bootstrap
- Gradio
- JavaScript
- Plotly
- Highcharts