01 Systems with evidence
Python & AI Developer | LLM Integration, Backend Systems & AI Automation
I build Python AI and backend systems that connect language models, APIs, data, and automation workflows, grounded in applied mathematics and mathematical modeling.
My current focus includes retrieval-augmented generation (RAG), Model Context Protocol (MCP), AI agents, and reliable LLM integration.
02 Selected work
One explores low-latency voice interaction. The other makes domain rules inspectable and versioned. Both are presented with their current limits intact.
FastAPI · WebRTC · Gemini Live
A local real-time voice demo connecting a browser microphone to an external AI service through a Python backend and bidirectional WebRTC audio.
Read the case filePython · JSON Schema · rules-as-code
An open-source rules-as-code project for auditable Shariah stock screening, designed around versioned rules, explicit uncertainty, and sourced evidence.
Read the case file03 Throughline
Make the path between browser, backend, data, and external services visible.
Separate working scope from planned capability, especially in AI and finance-adjacent software.
Keep rules, inputs, versions, and source references close to the result they produce.
04 Find me online