AI Infrastructure & Software Engineer
AI Infrastructure · Backend · Full-Stack Systems
I build AI systems and the infrastructure that runs them. My recent work spans LLM inference optimization, retrieval augmented generation, on-device AI, and multi agent workflows, backed by 2+ years shipping production web, mobile, and cloud native systems. I care about latency, throughput, and reliability, and I like owning products end to end, from frontend and backend APIs down to GPU level model serving and deployment.
Open to full-time software engineering roles starting January 2027
Skills
My experience spans model serving and applied AI through production APIs, cloud infrastructure, real-time systems, and cross-platform applications.
Serving and optimizing LLMs for throughput, latency, and memory efficiency.
RAG pipelines, retrieval, and multi-agent systems over real data.
Production APIs, real-time services, auth, and system design.
Containers, orchestration, CI/CD, and cloud-native deployment.
Cross-platform product interfaces on web, iOS, and Android.
Relational and NoSQL data, caching, and query optimization.
Background
ServBeyond Solutions
Runara
Svipes
Xelpmoc Design and Tech Limited
IIIT Hyderabad
Suntek Organization
Selected Work
A mix of on-device AI, full-stack products, and real-time systems — spanning inference, backend, cloud, and mobile.
Privacy-first React Native assistant that performs text and image Q&A entirely on-device using Qwen3-VL through llama.rn. It includes local model downloading and verification, streamed responses, persistent conversations, cancellation, and memory-aware failure handling without sending user content to a server.
AI powered care transition platform that turns discharge PDFs into structured clinical data and actionable care plans. Built with FastAPI, React, PostgreSQL, and Google ADK, using multi agent validation and confidence threshold guardrails to catch post discharge risks earlier.
Multi agent code analysis platform that ingests GitHub repositories or ZIP files and returns architecture insights, API understanding, and context aware Q&A. Built with FastAPI, LangChain, AutoGen, FAISS, and OpenAI, using a RAG layer over the codebase and specialized SDE, PM, and QA agents.
Team project · Backend and cloud contributor
Cloud native, serverless analytics platform on AWS using Lambda, EKS, S3, DynamoDB, SQS, API Gateway, and CloudFormation. Microservices auto scale from 2 to 10 pods behind a load balancer, with infrastructure as code and a GitHub Actions CI/CD pipeline that passed security scanning with zero critical vulnerabilities.
Team project · Full-stack and infrastructure contributor
Full stack campus event platform with CAS SSO, JWT role based access control, Stripe payments, and waitlist automation. Built with React, FastAPI, and MongoDB and deployed on Kubernetes with CI/CD, holding sub 200ms API performance under 100+ concurrent users.
Local AI image generation system that strengthens weak prompts through an LLM powered enhancement layer before diffusion inference. Built with FastAPI, llama.cpp, and diffusers to combine local model serving, prompt refinement, and modular backend orchestration without relying on hosted APIs.
Contact
If you are hiring or want to talk through a problem in AI inference, backend systems, or full-stack product work, I would love to hear from you.
Whether it's a full-time role, a technical discussion, or a collaboration, feel free to reach out.
AI infrastructure and LLM inference, applied AI and agents, backend and APIs, cloud-native systems, and full-stack and mobile product work.