- Local-first chat experiments (Convo-AI) that run FastAPI + Ollama on my laptop.
- Website embeds (ProjectHub) that pull from documentation to answer basic questions.
- Prompt libraries + README honesty logs that show how much AI wrote vs. what I edited.
- No production integrations, no enterprise telemetry—just learning projects with clear TODOs.
AI & Automation Engineer
Practicing AI + automation with transparent limits
I haven’t launched AI copilots for paying customers. These prototypes document how I learn with ChatGPT, Copilot, and local LLMs.Current focus
Proof on GitHub
Convo-AI
- FastAPI backend + simple UI for local chat flows.
- Uses Ollama models and environment variables documented in the repo.
- Disclosure: AI wrote the first draft of most endpoints; I kept prompts + edits in the README.
ProjectHub Copilot
- Express proxy + lightweight widget that surfaces answers from my own case studies.
- Currently supports a single route and manual deployments.
- Backlog items (usage analytics, guardrails) are documented as future work.
What I’m experimenting with
Python + FastAPI (learning)Node.js / Express (comfortable for prototypes)LangChain (exploring)Ollama + local LLMsOpenAI / Anthropic APIsSupabase (learning for vector stores)GitHub Actions for small deploys
Every repo labels what’s working vs. what’s aspirational so collaborators know the maturity level.
What I still need to learn
- Responsible AI guardrails (policy checks, escalation paths) in production environments.
- Measuring ROI beyond “this feels faster on my laptop.”
- Scaling prompt orchestration with queues, storage, and audit requirements.
- Security/privacy reviews for AI features before they reach real users.
If you mentor junior engineers on applied AI/automation, I’d appreciate pairing sessions.