Building reliable backend systems, AI agents, and scalable infrastructure.
I am focused on designing systems where AI is powerful but observable—ensuring that real-world operations remain bounded by clear, deterministic rules.
|
🏗️ Building Reliable retrieval pipelines (RAG), web-search fallbacks, and tool-using AI workflows. |
⚙️ Architecting Low-latency, scalable APIs with Python (FastAPI) and Node.js. |
|
🛡️ Systems Engineering Exploring idempotency, UDP networking, rate limiting, and zero-trust security. |
🚀 Improving Advanced DSA, distributed systems architecture, and system design. |
Technologies I work with daily, organized by domain.
Backend & Systems:
AI & Machine Learning:
Databases & Infrastructure:
| Project | Description | Technologies |
|---|---|---|
| 📚 Corrective RAG Analyzer | Academic paper analyzer using corrective RAG, retrieval grading, web-search fallback, and hallucination checks. | Python, FastAPI, LangGraph, FAISS |
| 🛡️ VILR | Zero-trust authorization gateway for agent-driven payments, separating AI intent from deterministic payment approval. | Python, Redis, Docker, OPA |
| 🤖 BossConsole | Multi-platform harness for AI agents featuring a native, multi-threaded operator's console with real browser/terminal access. | JVM, AI Agents, MCP Tools |
| ⚙️ UDP Media Delivery | Reliable real-time UDP media delivery implementation addressing packet loss and delay constraints. | C, Python, Makefile |
| 🚦 Rate-Limiter | Systems engineering project exploring idempotency, rate limiting, and performance constraints. | Python |