Senior engineer working across AI platforms, analytics engineering, and production data systems.
I build Python and cloud-native systems that help teams move from prototypes to reliable, usable AI and data products. My recent work has focused on conversational analytics, LLM agents, evaluation pipelines, Google Cloud deployment, and infrastructure-as-code.
- Production AI and data systems on Google Cloud, including Vertex AI, Cloud Run, and Terraform-managed infrastructure
- LLM agent development, evaluation, benchmarking, and iteration workflows
- Python tooling for data integration, API clients, and AI platform development
- Analytics engineering with dbt, Snowflake, BigQuery, and robust CI/CD practices
- Teaching, mentoring, technical writing, and helping teams adopt better engineering practices
A modular text-to-SQL toolkit built as a Python package. It includes logical planning, schema inspection, semantic linking, and database connector support across BigQuery, Snowflake, DuckDB, MotherDuck, and SQLAlchemy-compatible systems.
Video walkthrough: piglets text-to-SQL planning
Relevant areas: Python packaging, LLM application architecture, database tooling, provider abstractions, tests, examples, and open-source project structure.
An Apache Beam pipeline for extracting and transcribing frequently released newscasts using OpenAI Whisper. The project includes Docker, Cloud Build, Terraform, tests, and a portable pipeline structure for parallel processing.
Video walkthrough: newscast transcriber
Relevant areas: data pipelines, batch processing, containerisation, infrastructure-as-code, CI, and cloud deployment.
A set of Google ADK example projects that build up from a first agent to tool-using agents, distinct sub-agents, and sequential agent patterns.
Video walkthrough: Google ADK example series
Relevant areas: agent design, Google ADK, progressive learning resources, and practical AI engineering examples.
A minimal DuckDB-Wasm implementation using vanilla HTML and JavaScript.
Video walkthrough: DuckDB-Wasm in the browser
Relevant areas: browser-based analytics, local-first data tooling, and simple public learning resources.
A Neovim plugin for SQL transpilation built on SQLGlot.
Relevant areas: developer tooling, SQL, Lua, and workflow automation.
Python, SQL, Bash, YAML, GCP, Vertex AI, Cloud Run, Terraform, Docker, Git, GitLab CI/CD, Azure DevOps, dbt, Snowflake, BigQuery, Pytest, Streamlit, Apache Beam, LangChain, Google ADK, Gemini, OpenAI, DuckDB.
My background spans analytics engineering, data science, software delivery, and teaching. I have built Python data pipelines, developed AI and data tooling, delivered workshops, mentored engineers, and presented technical work to product and platform teams.