I’m a Systems Engineer at VaultIQ.ai, working at the intersection of AI systems, backend engineering, retrieval infrastructure, and LLM-powered products.
My work focuses on building practical AI systems that move beyond simple prompting — systems involving document ingestion, chunking, embeddings, retrieval, reranking, knowledge graphs, evaluation, and production-ready APIs.
I enjoy designing systems where LLMs, databases, search, and backend infrastructure work together reliably.
- 🔭 Currently working on AI-native knowledge retrieval systems
- 🧠 Interested in RAG, Graph RAG, agentic workflows, search, and LLM evaluation
- 🛠️ Comfortable across Python, backend APIs, databases, Docker, Linux, and ML tooling
- 🚀 Building towards deeper expertise in distributed systems, GPU programming, and scalable AI infrastructure
June 2026 - Present
Working on AI-driven systems for knowledge retrieval and automation.
Key areas I work around:
- Retrieval-Augmented Generation systems
- Document ingestion and indexing pipelines
- Chunking strategies and contextual retrieval
- Embedding pipelines and vector search
- Knowledge graph based retrieval
- Backend APIs for AI products
- Evaluation and debugging of LLM pipelines
- Infrastructure for deploying and scaling AI systems
December 2025 - May 2026 Worked on GyanSetu, an AI-powered learning and teaching platform focused on making educational content more structured, searchable, and interactive. Key areas I worked on:
- Built LLM-powered learning workflows for educational content
- Worked on ingestion and structuring of curriculum-level study material
- Designed retrieval pipelines for question answering and teaching assistance
- Explored knowledge graph based representation of topics, examples, questions, diagrams, and prerequisites
- Worked on teacher-facing and student-facing AI assistant ideas
- Contributed to experiments around Graph RAG, content understanding, and pedagogical structure
📆 2021 - 2025 📍 Indian Institute of Technology, Madras
- LLM applications and RAG systems
- Embeddings, vector search, reranking
- Graph-based retrieval and knowledge graphs
- Scikit-learn, Pandas, NumPy
- PyTorch, TensorFlow
- LangChain / LangGraph-style workflows
- LLM evaluation and pipeline debugging
- Obervability Stacks (LangSmith/Langfuse)
- Python, FastAPI, Flask
- Node.js, Express.js
- REST APIs and service design
- Docker, Linux, Nginx
- Background workers and async workflows
- API integration and automation systems
- PostgreSQL
- pgvector
- SQL
- MongoDB
- Redis
- Graph-style querying and retrieval workflows
- React
- Vue.js
- JavaScript / TypeScript
- HTML, CSS, Tailwind CSS
- Git / GitHub
- Docker
- Postman
- Linux servers / VPS deployment
- OpenRouter and LLM APIs
- Cloud and GPU-based experimentation
- Building practical AI systems, not just demos
- RAG pipelines that can be evaluated and debugged
- Backend systems that make LLM products reliable
- Knowledge graphs and structured retrieval
- Developer tooling, automation, and infrastructure
- Learning lower-level systems concepts over time
-
📚 Reading fiction and non-fiction Sci-fi, fantasy, classics, philosophy, psychology, and history
-
🗣️ Learning languages Currently learning German 🇩🇪
