My approach
Practical
AI where it adds value, not everywhere. No hype, but results.
Transparent
Clear about what AI can and cannot do. Realistic expectations.
Privacy-aware
Data processing with attention to GDPR and business-sensitive information.
Maintainable
Solutions that still work and can be adjusted a year from now.
What I build
LLM Integrations
Integrate OpenAI, Claude, or open-source models into your application for text, analysis or conversation.
RAG Systems
AI that bases answers on your documents and data. Knowledge bases, support bots, document analysis.
Prompt Engineering
Design effective prompts for consistent, reliable AI output in production environments.
AI Workflows
Automate complex tasks by combining AI with traditional software. Classification, extraction, summarization.
Evaluation & Guardrails
Monitor and control AI output. Quality control, content filtering, fallback logic.
Fine-tuning Advice
When to fine-tune or not? Cost-benefit analysis and guidance on model adjustments.
Technologies
Frequently asked questions
Is AI suitable for my use case?
What about the costs of AI APIs?
Is my data safe?
What if the AI makes mistakes?
Deploy AI?
Tell me about your situation and I'll think along about the possibilities.