Building Intelligent Systems
Bridging the gap between cutting-edge AI research and real-world applications through RAG pipelines, multi-agent systems, and applied ML.
M.Sc. in Data Science, Christ (Deemed to be University). Specialized in LangChain, RAG, vector search, and FastAPI, with 2 years of prior full-time experience in SQL/ETL data engineering bringing production-grade reliability discipline to AI system design.
Work Experience
Professional journey spanning AI/ML engineering and data engineering across an AI-focused startup and a global consulting firm.
Vision AI Engineer (Graduate Practicum)
- Automated employee/visitor identification and attendance logging by fine-tuning an OpenCV facial recognition pipeline, deployed live via a FastAPI backend and Next.js/React dashboard (Postgres, Docker, VPS, Tailscale-tunneled RTSP)
- Piloted the fine-tuned pipeline on 2 cameras in production and led investigation into GPU-efficient scaling strategies (quantization, batching) to extend the system to the full 40-camera deployment
AI/ML Engineer (Graduate Practicum)
- Designed a 3-agent LangChain/Google ADK workflow with a FastAPI orchestration layer, automating a multi-step enterprise process (retrieval, routing, response generation) as a proof-of-concept for scalable automation
- Improved contextual answer relevance by building and evaluating a RAG pipeline (FAISS vector database, Google Gemini embeddings) over PDFs and chat history, benchmarking retrieval quality against labeled queries
- Explored low-code agent creation patterns to simplify how non-technical users could configure and deploy custom agents within the POC ecosystem
- Prototyped orchestration patterns for coordinating multiple specialized AI agents toward a shared task
Associate Analyst – Data/ETL Engineering
- Protected the accuracy of 80+ SAP BI reports by validating 184 ETL jobs, reviewing SQL (stored procedures, DML, DDL) for correctness at production scale
- Root-caused and resolved 5 critical ETL/SQL defects through workflow redesign, improving downstream reporting stability
- Built an automated PDF validation proof-of-concept that cut manual QA effort by 60% — foundational data-reliability experience now applied to validating AI/RAG pipeline outputs
Featured Projects
RAG systems, multi-agent architectures, and applied ML — spanning agriculture, finance, energy, and generative AI.
KrishiSahayak – AI Agri Scheme Assistant
A RAG-based chatbot helping Indian farmers understand government agricultural schemes quickly and accurately. Combines local document retrieval (ChromaDB) with live web search fallback (Serper, DuckDuckGo) for grounded, cited answers, with multi-provider LLM support (Gemini, OpenAI, Groq) and configurable response modes.
Agricultural Multi-Agent System & Research
A 4-agent Agentic AI system (advisory, pricing, pest, chat) with an LLM coordinator and FastAPI backend for farmer queries. Combines RAG over government scheme PDFs, price forecasting (ARIMA/SARIMA/Prophet), and a 14-class pest detection CV model. Co-authored paper, "A Farmer-Centric Orchestration Framework for Collaborative Multi-Agent Agricultural AI Systems," presented at the 3rd World Congress on Smart Computing (WCSC 2026), Bangkok, Thailand — January 2026.
More Projects
Financial advisory system using specialized AI agents (DSPy + LangChain) for investment advice, market analysis, and portfolio insights via a chat interface. FastAPI + PGVector backend, React/Vite frontend.
Fine-tuned Stable Diffusion (LoRA/PEFT) generating custom shoe designs from text prompts, deployed as a Gradio app hosted on Hugging Face Spaces.
Variational Autoencoder that reconstructs uploaded face images and generates new random faces from a learned latent space, with an interactive Gradio interface.
Interactive app forecasting electricity demand from climatic data, comparing Prophet, SARIMAX, and LSTM models side by side under different weather scenarios.
ML system identifying and classifying malware in IoT network traffic using Random Forest and XGBoost, with a feature engineering pipeline to reduce false positives.
Full-stack event management app with a Flask backend, MySQL database, and responsive HTML/CSS/JS frontend, including authentication and CRUD APIs.
Skills & Technologies
Core strength in RAG pipelines, multi-agent LLM systems, and vector search, backed by a data engineering foundation.
AI / LLM Engineering
ML / Computer Vision
Backend & APIs
Data & BI
Infrastructure & Languages
Education
Master of Science in Data Science
Bachelor of Science in Computer Science, Mathematics & Electronics
Certifications
Let's Connect
Open to discussing new opportunities, collaborations, and AI engineering roles.
Available for Opportunities
I'm always interested in discussing new projects, roles, or opportunities to contribute to AI and data engineering teams. Whether you're looking to collaborate or just want to connect, feel free to reach out.
-
Email
-
Phone
-
LocationBengaluru, Karnataka
Online Presence
Connect with me on professional networks to stay updated with my latest projects and contributions to the AI and data science community.
-
LinkedIn
-
GitHub