Experience
- Build LLM-powered production systems end-to-end: API integration, prompt engineering, error handling, deployment
- Prototype LLM integrations for clients, moving from proof-of-concept to deployed internal tools
- Benchmark new models and frameworks against real use cases, not just leaderboard scores
Python
TypeScript
PyTorch
Anthropic API
PostgreSQL
Docker
GCP
Terraform
- Build end-to-end ML pipelines: data cleaning, feature engineering, model training, deployment
- Work directly with clients, no middle layers, translating business problems into technical solutions
- Delivered demand forecasting for retail, document classification for legal, LLM-powered text extraction from PDFs
Python
Pandas
Scikit-learn
TensorFlow
OpenAI APIs
SQL
Docker
Open Source Projects
Open Source
Batteries-included Pi package with always-on workflow for spec clarification, bounded-slice implementation, repo-map awareness, and skill discovery.
Open Source
Workflow layer on OpenCode adding durable project memory, plan-before-edit guard, TDD discipline, and automatic skill discovery.
Open Source
Chrome/Edge side-panel AI assistant for browser automation. Scrapes, fills forms, chains across tabs, exports to markdown/PDF. Local-only storage.
Education
Lancaster University Leipzig
Thesis: Compared Q-learning, DQN, and Dueling DQN against MCTS in a 4×4×4 Tic-Tac-Toe environment, investigating where neural function approximation outperforms classical tree search and vice versa.
Skills
AI & ML
LLM system design
Prompt engineering
OpenAI APIs
Anthropic API
TensorFlow
Scikit-learn
PyTorch
Programming
Python
TypeScript
Rust
Go
Java
C
Lua
Infrastructure
Docker
PostgreSQL
Linux
Git
CI/CD
Microsoft Certified: Azure AI Fundamentals (AI-900)
Anthropic API Certification
Languages
German Native
English C2
Romanian Fluent