// manifesto
One person instead of a distributed team: interface, API, database and deployment — in a single pair of hands. Nothing gets lost between departments, because there are no departments.
The foundation: readable, typed, optimised. Python-first thinking, SQL fluency, comfortable Java.
Interfaces that don't lag and don't fall apart. Responsive, accessible, pixel-faithful.
APIs, databases and deploys. From data schema to production healthcheck — no "not my area".
Cloud-native shipping on GCP & Azure. Containerised, monitored, versioned.
Production AI, not notebook demos: agentic pipelines, retrieval systems, evaluation harnesses and cost telemetry.
I'm a Computer Science graduate from Osmania University (M.Sc, 2026) with hands-on experience in full-stack web development, cloud platforms, and database management. Proven track record of developing scalable applications serving 500+ users and leading technical teams.
Skilled in Python, Java, Flask, Django, and cloud technologies with strong expertise in REST API development and database optimization. Just completed my M.Sc in Computer Science and am actively contributing to real-world projects.
Today I build AI systems that run in production — agentic RAG pipelines, evaluation harnesses and cost controls — not notebook demos. My main project, CORPUS, is an open-source agentic RAG system for research papers: hybrid retrieval on OpenSearch, self-correcting LangGraph pipelines, and nightly RAGAS evaluation.
Building and maintaining production web platforms for the organisation — landing pages, internal portals and automation. Backend-first work in Python, Flask, PostgreSQL, Redis, Docker and Nginx — the deployment experience behind the AI systems I ship today.
Architected Flask-based attendance tracking system managing 500+ users, reducing administrative processing time by 7 hours weekly. Developed automated SMS and email notification system processing 200+ daily alerts. Collaborated with an 8-member cross-functional team implementing JWT-based authentication.
Completed Master's degree with focus on Operating Systems, Advanced Java, Automata Theory, Computer Networks, Data Structures and Algorithms. Active member of React Club.
Graduated with GPA: 8.47/10.0. Relevant coursework: C, C++, Data Structures and Algorithms, Database Management Systems, Web Development.
Google Cloud Innovator • Microsoft Azure Fundamentals • Introduction to Data Science (Cisco) • Introduction to Cybersecurity (Cisco) • HackerRank Problem Solving (Basic) • Introduction to Large Language Models • Generative AI and Ethics in AI (LinkedIn Learning)
Complete web applications from frontend to backend with modern tech stacks. End-to-end, scalable.
Cloud migration, deployment automation, and infrastructure management. Certified on GCP & Azure.
Custom workflows, data processing, and integration solutions that streamline your operations.
Beautiful, user-centric interfaces with smooth animations and micro-interactions.
// numbers from projects and internships I have been involved with
Open-source agentic RAG for research papers — every claim cited to the exact PDF page. Hybrid BM25 + bge-m3 on OpenSearch, cross-encoder reranking, self-correcting LangGraph pipeline, RAGAS evals and cost telemetry.
ai · rag · langgraph · mcpRetrieval-augmented assistant over Kubernetes documentation — grounded answers with source citations instead of hallucinations.
ai · rag · retrievalOne SQL interface over APIs, files and live sources — built for agents. Ask in SQL, get answers from anywhere.
ai · agents · sqlKubeflow documentation AI agent built to power the Kubeflow website — accurate, version-aware docs answers.
ai · docs · agentsSecure donor↔NGO platform with real-time donation tracking, impact reporting and Stripe payments.
front · back · e2eRole-based attendance system for 500+ users with automated SMS/email alerts and analytics.
back · dashboard · alertsBrowser-based code editor replicating the VS Code experience with syntax highlighting.
front · editor · jsThis very site — brutalist redesign with full-stack backend, rate limiting and e-mail automation.
front · back · devopsRequirements, constraints and expectations. Then an honest answer on what's achievable.
A plan: stack, data schema, deployment strategy. No surprises later.
One-week iterations, a demo every Friday. Progress you can see.
Green checks, clean deploys. Releases without surprises.