I am an AI Systems Engineer and second-year dual-degree computer science scholar pursuing a BS in Data Science and Applications at IIT Madras alongside a B.Tech in Computer Science and Engineering at AMET University (CGPA: 9.52 / 10.0).

My core specialization spans Generative AI, Autonomous Agent Architectures, Abstract Syntax Tree (AST) Document Compilers, and Context & Token Optimization. I focus on the hard engineering bottlenecks of modern LLMs — eliminating chunk-boundary context fragmentation in RAG and slashing token consumption in iterative research agent swarms.

• Research & Engineering Focus: Agentic workflows, AST semantic parsing, production RAG pipelines, and token deduplication.
• Current Track: Maintaining open-source packages and scaling enterprise AI infrastructure.
• Location: Chennai, India (Available for high-impact AI Engineering & Research collaborations).

Autonomous Agent Context & Token Optimization Engine • Python, LangChain, AST

• Researched, architected, and published an open-source PyPI package that solves exponential token burn and API rate-limiting in iterative autonomous research agents.
• Engineered a semantic concept-diff deduplication algorithm that benchmarks newly gathered research nodes against existing agent memory, pruning redundant tokens before LLM context injection.
• Delivers up to 85% context token reduction per research cycle with zero synthesis degradation; provides drop-in modular adapters for LangChain.

AST-Driven Hierarchical Document Compiler for Advanced RAG • Python, Gemini API, AST, Docker

• Engineered an end-to-end compiler converting multi-format unstructured corpora (PDF, DOCX, Markdown) into structured Universal Open Knowledge Format (OKF) AST nodes.
• Eliminated the chunk-boundary semantic fragmentation and lost cross-references caused by naive character chunking in conventional RAG pipelines.
• Integrated the Google Gemini API with strict JSON schema enforcement to parse entity relationships into knowledge graphs, cutting retrieval latency by 40% while maintaining multi-hop context integrity.

_{May 2026 – Present | Chennai, India}

• Founded AI and digital transformation solutions firm; architected and shipped production AI automations and custom client platforms.
• Built resilient generative AI pipelines using Google Gemini & OpenAI APIs, LangChain, PostgreSQL, and Pydantic schema validation.
• Engineered backend middleware with real-time token tracking, rate-limiting, and deterministic JSON outputs.

_{Jun 2026 – Present | Remote}

• Engineered an intelligent WhatsApp conversational assistant deployed for hotel staff SOP guidance and real-time guest service automation.
• Integrated WhatsApp Business API webhooks with LLM backend pipelines, implementing intent classification and context routing to human staff for edge cases.

_{Mar 2026 – May 2026 | Remote}

• Integrated GenAI content synthesis and automated assessment generation into a production cloud Learning Management System (LMS).
• Developed responsive UI components connecting client-side applications to backend AI service endpoints, optimizing payload latency.