Systems, not tutorials.
End-to-end AI systems built with an infrastructure mindset — retrieval, agents, inference, and evaluation. Click any project for a full case study.
Quarry
Production AI systems platform evolving from retrieval and RAG into agents, memory, evaluation, observability, and inference infrastructure.
Aeroguard
Detects aircraft anomalies using PCA and DBSCAN on high dimensional NASA telemetry data with visualization and sensor level diagnostics.
OnboardAI
Agent AI workflow using planner executor validator architecture with LLM driven decision loops and automated orchestration.
An engineer for production AI.
I focus on the systems that make AI usable in the real world — retrieval, agents, evaluation, and inference. My work sits at the intersection of backend engineering and AI infrastructure.
Retrieval Architectures
Hybrid retrieval, embeddings, re-rankers, and vector stores tuned for real workloads.
Agent Workflows
Planner-executor-validator loops with tool-use, memory, and reliable orchestration.
Production Infrastructure
FastAPI, Redis, Postgres, Docker — designed for observability and resilience.
Evaluation Systems
Offline and online evals to make AI systems measurable and improvable.
Inference Serving
Latency-aware inference stacks and cost-efficient serving strategies.
Backend AI Engineering
From API design to async systems — engineering as a first-class citizen.
The path so far.
Started building AI systems
Began deep work on retrieval, embeddings, and LLM application design.
Hack For Tomorrow 2.0
Delivered a complete AI project at a national hackathon.
Building Quarry AI Platform
Designing an AI knowledge infrastructure platform with retrieval, memory, agents, evals, and inference.
Targeting AI Systems and Inference Engineering roles
Aiming for high-signal engineering roles focused on production AI infrastructure.
Selected highlights.
A snapshot of milestones from building end-to-end AI systems and competing at the national level.
Hack For Tomorrow 2.0
Delivered a complete AI system for a national hackathon.
Smart India Hackathon
Contributed to a nationally recognized engineering initiative.
Multiple complete AI systems
Shipped retrieval systems, agent workflows, ML anomaly detection, and CV inference systems.
The stack I ship with.
Engineered for high performance and production reliability.
AI Systems
Backend Infrastructure
Machine Learning
Deployment & Infrastructure
Let's build something serious.
Reach out for AI systems roles, collaborations, or engineering conversations about retrieval, agents, and inference.