AI Engineer · GenAI Architect · Software Developer
class Nitin:
def build(self, problem: Messy) -> System:
return retrieve(problem) | reason() | ship() # no demos. production.I build the layer between a model and something people can actually use. retrieval that returns the right chunk, agents that know when to stop, models tuned for the job.
01 RAG & Agents |
hybrid retrieval · multi-agent orchestration · guardrails that fail loudly |
02 MCP Servers |
custom tools & connectors · giving models real hands |
03 LLM Fine-Tuning |
LoRA/PEFT · domain adaptation · evals, not vibes |
04 Computer Vision |
detection · OCR pipelines · vision + language systems |
05 Voice AI |
streaming STT → LLM → TTS · sub-second turn-taking |
06 Full-Stack |
the API, the pipeline, the interface — shipped end to end |