Help & Supportcareers@gravityer.com
AI/ML Engineer with hands-on experience designing and shipping agentic RAG systems.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI/ML Engineer with hands-on experience designing and shipping agentic Retrieval-Augmented Generation (RAG) systems using LangGraph, hybrid retrieval, and Large Language Models (LLMs). Experienced in building end-to-end pipelines covering document ingestion, self-correcting query loops, and persistent memory architectures. Seeking Junior Al Engineer, GenAl Engineer, LLM Engineer, or RAG Engineer roles.
Periyar Maniammai Institute of Science and Technology
B.Tech · Computer Science
N/A – June 30, 2024
Agentic Document Q&A System
June 24, 2026 – Present
Architected a Corrective RAG agent using LangGraph with conditional routing across router, retriever, grader, hallucination checker, and query rewriter nodes. Implemented semantic chunking (meaning-based splits) and parent-child chunking (300-character child chunks for precise retrieval, 1500-character parent chunks for LLM context). Built hybrid retrieval combining BM25 keyword search and ChromaDB vector search; merged and reranked results using cross-encoder/ms-marco-MiniLM-L-6-v2. Designed a self-correcting loop in which the retrieval grader filters irrelevant documents and automatically rewrites and retries the query when too few relevant results remain. Added a hallucination checker that validates generated answers against retrieved context before returning a response to the user. Built a Streamlit UI with a live reasoning trace showing each LangGraph node decision in real time; supports dynamic PDF upload.
View ProjectLLM Research Agent
June 24, 2026 – Present
Built a 4-node LangGraph pipeline (search, store, retrieve, synthesize) for persistent web research memory. Implemented query-scoped retrieval with metadata filtering to prevent topic pollution across research sessions. Integrated the Tavily web search API and Groq Llama 3.1 8B model to generate fast, sourced answers backed by persistent ChromaDB storage.
View ProjectCultural Fit Analysis
The candidate's projects are highly aligned with the 'Generative AI Engineer' target role, showcasing deep engagement with relevant technologies and methodologies. The focus on personal projects demonstrates initiative and passion for the field. However, the lack of diverse project types, team experiences, or professional work experience limits the assessment of broader cultural fit and adaptability to different team dynamics or business contexts. The candidate is a recent graduate, which naturally limits the breadth of experience.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a strong problem-solving orientation and an ability to translate complex AI concepts into functional systems. The detailed breakdown of architectural decisions and implementations suggests good analytical skills. However, without direct interaction or psychometric test results, it is difficult to assess communication, teamwork, or stress handling capabilities.