Experience & expertise.

AI & evaluation

Agent and evaluation harnesses, benchmarking, LLM-as-judge, hallucination detection, NLP, RAG, multi-agent orchestration, MCP/tool calling, model training and fine-tuning, reinforcement learning, inference optimization and observability.

Engineering

Python, TypeScript, JavaScript, Java, C, C++, SQL, Bash/Linux; React, FastAPI, REST APIs, async Python, WebSockets, PostgreSQL, Redis, SQLAlchemy, Pydantic and pytest.

Infrastructure

AWS, Terraform, Docker, Kubernetes, CI/CD, Azure ML, Google Cloud and Databricks.

Machine learning

PyTorch, Hugging Face, scikit-learn, XGBoost, Pandas, NumPy and PySpark.

Quantitative methods

A/B testing, experimental design, causal inference, Bayesian statistics, hypothesis testing, power analysis, bootstrap confidence intervals, regression, time series and optimization.

Tools & frameworks

Pipecat, LangGraph, OpenAI Agents SDK, Claude Code, Codex and Cursor.

Certifications

AWS Certified Cloud Practitioner : May 2026

Anthropic Claude Certified Architect – Foundations : April 2026

Languages

German : fluent

Chinese (Mandarin) : conversational