Pamela Fox Portfolio
Principal Cloud Advocate at Microsoft
Principal Cloud Advocate at Microsoft focused on Python with Azure, Foundry, GitHub, and VS Code, which in practice means a steady run of RAG, MCP and agent talks, including five sessions at AI Engineer World's Fair 2026; background spans founding engineering at the Woebot therapy chatbot, curriculum work at Khan Academy, and academic teaching at UC Berkeley — her site indexes talks, interviews, and a public reading list alongside her writing.
Location
Unlisted — probably wherever the wifi is strongest.
Companies
Credentials
- USC — bachelor's & master's in Computer Science, minors in 3D Animation and Linguistics
Skills
Notable
- Lectured at UC Berkeley, teaching CS61A and CS302
- Built Proficient Python, a free interactive online Python course
- Contributor to Azure-Samples/azure-search-openai-demo, a Python RAG chat sample with 7.8k GitHub stars
- Speaker at AI Engineer World's Fair 2026 (five sessions listed on her talks page)
What makes it work
A breakdown of the choices that make this portfolio stand out.
A career history that traces directly to current applied-AI work
Founding engineering at an actual production chatbot (Woebot) predates the current LLM wave and gives her current RAG/AI advocacy work a track record grounded in a shipped, real-world conversational product.
Multiple content formats indexed from one homepage
Talks, interviews, blog posts, and a reading list are each given their own section rather than mixed into a single feed, making years of scattered public work easy to navigate from one place.
Honest, stated constraints on availability
Explicitly stating limited time due to family and a full-time job, rather than implying unlimited availability, sets realistic expectations for anyone reaching out — a small detail that builds trust.
What AI Engineers can take from this
Specific, actionable tips to apply to your own portfolio — no generic advice.
If your current AI work builds on earlier, less trendy experience (like an older chatbot or NLP project), state that lineage explicitly — it shows real depth, not a recent pivot.
Index different content types (talks, interviews, writing) into separate sections from one homepage rather than one undifferentiated blog feed.
State your real availability and constraints plainly rather than implying you're always open — it sets honest expectations upfront.
Projects
The projects page opens by admitting most of its entries are "dead/defunct/out of order/in a better place", and lists them anyway, each one labeled. Pamela Fox keeps a portfolio as an honest ledger rather than a highlight reel, which makes the live projects easier to trust.
RAG chat app with Azure OpenAI and Azure AI Search (azure-search-openai-demo)
A ChatGPT-like frontend over your own documents using retrieval-augmented generation. Azure AI Search handles indexing and retrieval and Azure OpenAI generates the answers. The backend is Python, with JavaScript, .NET and Java variants built from it. It deploys with the Azure Developer CLI and ships with a productionizing guide and security notice.
It is a reference RAG architecture that thousands of developers fork as a starting point. Contributing to it means working on ingestion, retrieval, streaming and deployment at the level that production teams copy.
End-to-end RAG: document ingestion, indexing, retrieval, grounded answers
One-command deployment with the Azure Developer CLI
Used as the demo code behind her RAG talks
Proficient Python
A free online course for building real confidence in Python. It covers variables and functions, control flow, loops and lists, strings and dictionaries, file handling, and object-oriented programming, through articles with inline quizzes, in-browser coding exercises with an integrated test runner, and unit projects in Colab notebooks.
It carries her Berkeley CS61A teaching into a self-serve product: scaffolded starter code, video walkthroughs and auto-graded exercises. That is curriculum design and engineering in one.
Coding exercises run and grade in the browser
Four units, from fundamentals to OOP with inheritance and composition
Open-ended unit projects in Jupyter notebooks on Colab
Recursion Visualizer
An online tool that draws the call graph of a Python function, especially one with multiple recursive calls. Pick an example (virfib, count_partitions, luhn_sum) or paste your own def and function call, then visualize.
It is a small teaching tool aimed at the exact topic students struggle with in intro CS, and it shows an educator turning a classroom pain point into software.
Paste any recursive Python function and see its call tree
Preloaded CS61A-style examples like count_partitions
Runs in the browser
More AI Engineer portfolios
Ready to build your portfolio?
Follow the AI Engineer roadmap — skills, projects, and timeline to get hired.