Generative AI Engineer with less than a year in LLM Applications & 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
Generative AI Engineer with hands-on experience building production-ready LLM applications, Retrieval-Augmented Generation (RAG) systems, and multimodal AI solutions. Delivered 3 end-to-end deployed AI systems across legal, media, and multimodal domains. Skilled in LangChain, FAISS, ChromaDB, HuggingFace, and Docker-based deployment. Experienced in agentic AI workflows, prompt engineering, and scalable cloud deployment. Actively seeking Generative AI / LLM Engineer roles (0-2 yrs).
IIT-M GUVI
Advanced Programming Professional & Master Data Science · Large Language Models (LLMs), RAG Systems, MLOps, Generative AI, Deep Learning, Data Engineering, Python Programming
February 1, 2024 – October 1, 2024
Paavai College of Engineering
B.E. Electronics & Communication Engineering · Data Structures & Algorithms, Python Programming, Signal Processing, Communication Systems, Microprocessors
August 1, 2019 – April 1, 2024
Legal Multi-Agent RAG System
October 1, 2025 – June 1, 2026
Architected a multi-agent RAG pipeline for legal Q&A; using Llama-3, LangChain orchestration, and a citation validation agent. Indexed and retrieved across 12+ legal documents using FAISS vector store and ChromaDB with metadata filtering, achieving sub-3s query response. Improved answer relevance by ~35% through optimised chunking strategy and semantic metadata filtering vs. naive retrieval baseline. Reduced LLM hallucinations by ~40% by implementing a grounding-based citation validation layer before response generation. Deployed interactive Gradio web interface enabling real-time legal querying with source citation display.
View ProjectPalludagam AI – Multimodal LLM System
June 1, 2025 – June 1, 2026
Built production-ready multimodal AI app processing 3 input types (image, audio, PDF) via Gemini 2.5 Flash API integration. Reduced Docker image size by ~40% using python:3.12-slim base, cutting cold-start time and cloud deployment costs. Automated CI/CD pipeline via Docker Hub (rajann71/palludagam-ai-hub:v1), enabling one-command reproducible global deployment. Designed real-time Streamlit UI with consistent cross-environment performance and zero "works on my machine" issues.
View ProjectGPT-2 Fine-Tuning & Deployment
March 1, 2025 – September 1, 2025
Fine-tuned GPT-2 transformer model on a custom domain dataset achieving stable convergence within 5 epochs and ~30% improved text generation quality. Built custom tokeniser and preprocessing pipeline; optimised hyperparameters (learning rate, batch size, warmup steps) for best validation loss. Deployed production-ready text generation web app on Streamlit with real-time inference and user-configurable generation parameters.
View ProjectAI-Powered Daily News Brief Generator
January 1, 2025 – July 1, 2025
Built transformer-based summarisation system processing 50+ articles/day across 5 topic categories with personalised filtering. Integrated 3 real-time external news APIs to stream and process live data with automated deduplication and relevance ranking. Deployed interactive Streamlit application with topic-preference UI, reducing reading time by an estimated 70% per user session.
View ProjectSingapore Resale Flat Price Prediction
November 1, 2024 – May 1, 2025
End-to-end regression ML pipeline; feature engineering, model validation, and Streamlit deployment. Achieved R2 score of 0.91 on test set.
View ProjectIndustrial Copper Price & Sales Modelling
October 1, 2024 – April 1, 2025
Random Forest regression for industrial price forecasting; outlier treatment, skew correction and feature selection — improved RMSE by ~22% vs baseline.
View ProjectPhonePe Pulse Data Visualisation
September 1, 2024 – March 1, 2025
Analysed 1M+ transaction records; EDA, trend analysis, and business intelligence dashboards for payment pattern insights.
View ProjectCultural Fit Analysis
The candidate's project portfolio is highly aligned with the 'Generative AI Engineer' target role, showcasing a deep interest and practical application in LLMs, RAG, and multimodal AI. The diversity of projects (legal Q&A, news brief, multimodal app) indicates adaptability and a broad interest within the AI domain. The academic background, including a Master's specialization in LLMs and Generative AI, further reinforces this fit. However, the lack of professional experience means cultural fit in a corporate setting is yet to be proven.
Soft Skills & Operational Fit
The candidate's project descriptions highlight a results-oriented approach, focusing on quantifiable improvements (e.g., reduced Docker image size, improved answer relevance, reduced hallucinations, improved RMSE). The emphasis on 'production-ready' and 'real-time' applications suggests an understanding of operational requirements and user experience. The academic nature of all projects means direct professional soft skill demonstration is limited, but the project outcomes imply problem-solving and execution capabilities.