Help & Supportcareers@gravityer.com
Generative AI Engineer with 8+ years in multi-agent AI systems and RAG pipelines.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Applied AI / Generative AI Engineer with 7.5+ years of total software engineering experience, including 2+ years building production-grade AI applications. Specialized in designing multi-agent AI systems, Retrieval-Augmented Generation (RAG) pipelines, and LLM-based solutions using LangChain and LangGraph. Brings a strong backend engineering foundation in Java, Spring Boot, microservices, and Kafka from the banking and payments domain, with hands-on experience integrating ML/LLM inference into scalable, production-ready systems. Focused on system design, explainability, and reliable end-to-end deployment of AI solutions.
The University of Texas at Austin
Post Graduate Program in Artificial Intelligence and Machine Learning · Artificial Intelligence and Machine Learning
March 1, 2025 – April 1, 2026
SRM University
Master of Computer Applications · Computer Applications
N/A – June 30, 2017
University of Calcutta
Bachelor of Science in Mathematics · Mathematics
N/A – June 30, 2014
Tata Consultancy Services
Senior Software Engineer
July 1, 2022 – Present
Chennai, Tamil Nadu, India
Verizon Data Services
Software Engineer
April 1, 2020 – July 1, 2022
Chennai, Tamil Nadu, India
Cognizant Technology Solutions
Software Engineer
December 1, 2017 – April 1, 2020
Chennai, Tamil Nadu, India
Banking Intelligence Assistant - Multi-Agent GenAI System
June 29, 2026 – Present
Designed and developed a multi-agent GenAI system using LangGraph to dynamically route user queries between a RAG-based policy agent and an LLM-powered SQL data agent, exposed via a FastAPI backend. Built the initial RAG pipeline using FAISS and HuggingFace embeddings, then migrated the vector store to Pinecone to simulate a production-ready architecture with managed storage, metadata filtering, and improved scalability. Improved retrieval relevance from ~75% to ~90% on an internal evaluation set by implementing query rewriting and Cross-Encoder reranking, and incorporated MMR retrieval to reduce redundant chunks and increase context diversity. Built a custom LCEL-based SQL chain to convert natural language queries into executable SQL for structured data retrieval, replacing an unreliable ReAct agent implementation. Developed an intelligent query-routing mechanism that improved response relevance and accuracy by directing queries to the correct agent based on intent. Enabled source attribution with clickable document references, increasing transparency and reducing hallucinated outputs. Deployed an interactive Streamlit application on Hugging Face Spaces for real-time public access and demonstration. Designed system architecture combining unstructured (documents) and structured (database) data sources, simulating real-world enterprise AI systems.
View ProjectAI-Powered Medical Assistant - RAG-Based Solution
June 29, 2026 – Present
Designed a Retrieval-Augmented Generation system using LangChain, HuggingFace embeddings, and FAISS to enable semantic search over a 4,000+ page medical reference corpus, built a conversational GPT-based interface for context-grounded responses, and deployed an end-to-end Streamlit + Flask microservice architecture for scalable inference.
TeamSync AI - AI-Powered Team Productivity Assistant
June 29, 2026 – Present
Built a solo end-to-end GenAI application (Build with AI Hackathon) featuring AI-powered task management, automated standup assistance, team chat, and meeting summarization. Integrated Groq's Llama-3.3-70B-Versatile model to power natural-language task creation, standup summarization, and contextual team chat, with session-state management in Streamlit for consistent multi-user workflow state. Configured Docker containerization and deployed the application to Google Cloud Run, resolving port-binding and container startup issues for production readiness. Redesigned the application UI into a professional dark-themed dashboard, improving usability and visual clarity for end users.
End-to-End MLOps Sales Forecasting Pipeline
June 29, 2026 – Present
Built a complete ML workflow covering data ingestion, preprocessing, model training, cross-validation, hyperparameter tuning, and evaluation for sales forecasting. Containerized the application using Docker, implemented CI/CD with GitHub Actions, and deployed an interactive inference application via Streamlit on Hugging Face Spaces.
"WINGS" Badge
TCS
June 1, 2026 – Present
Best Case Study Award and (S)miles Award
TCS Hackathon, Tata Consultancy Services
June 1, 2026 – Present
Python PCEP (Certified Entry-Level Python Programmer)
The OpenEDG Python Institute
June 1, 2026 – Present
"Kudos Rock Star Rookie" Award and Full Stack Java Developer Certification
Verizon
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's diverse project portfolio, including multi-agent GenAI systems, RAG solutions, and MLOps pipelines, indicates a strong alignment with an innovative and AI-focused culture. Their experience in both traditional enterprise software (banking) and cutting-edge AI development shows adaptability. The pursuit of a Post Graduate Program in AI/ML further highlights a commitment to continuous learning and growth, which is a positive cultural indicator.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through complex project implementations (e.g., query rewriting, multi-agent routing). Their experience in cross-functional team collaboration and leading CI/CD initiatives indicates good teamwork and operational leadership. The detailed project descriptions suggest a methodical approach to development and a focus on practical, scalable solutions.