βββββββ βββββββββββββββ βββ ββββββββββββββββββββββββββ βββββββ
ββββββββββββββββββββββββββββ ββββββββββββββββββββββββββββββββββββ
βββ βββββββββ ββββββββ βββββββ ββββββ ββββββ βββββββββββ βββ
βββ βββββββββ ββββββββ βββββ ββββββ ββββββ βββββββββββ βββ
βββββββββββββββββββ βββ βββ βββ βββββββββββ βββββββββββ
βββββββ βββββββββββ βββ βββ βββ βββββββββββ ββββββββββ
Math β Data β ML β LLM/AI Engineer
Mathematics graduate turned LLM Engineer. I build AI systems that survive contact with real-world data.
- π Deep in RAG pipelines, AI Agents & Agentic workflows right now
- π§ͺ Experimenting with model distillation and inference optimization
- π‘οΈ Red teaming LLMs to find where they break β before users do
- βοΈ Care about LLMOps: deployment, monitoring, cost efficiency
- π Classic ML & Deep Learning still have a place in my heart
- ποΈ Occasionally building data pipelines because good models need good data
LLM & AI
ML / DL
Infrastructure & MLOps
Vector DBs
Languages
Detailed breakdown of commits, PRs, reviews, and more β powered by lowlighter/metrics
current_focus = {
"RAG": ["advanced chunking", "hybrid search", "re-ranking"],
"Agents": ["multi-agent orchestration", "tool use", "memory systems"],
"LLMOps": ["inference optimization", "cost monitoring", "eval pipelines"],
"Red Team": ["prompt injection", "jailbreak patterns", "safety evals"],
}(coming soon β pinning the ones worth showing)
| Project | Stack | What it does |
|---|---|---|
| π AI Learning Platform | React + Vite + FastAPI | Interactive platform for learning AI/ML concepts |
| π Portfolio Web | Next.js + TypeScript | Personal site with terminal aesthetic |
| π More in repos β | β | Experiments in RAG, agents, and ML |
"Good models deserve good infrastructure."
Let's build something real.