Headline

AI Engineer | LLMs | RAG | MCP | Agentic AI | Prompt Engineering | Databases | Fine-Tuning | MLOps | FastAPI | Python | B.Tech NIT Warangal 2024 | Impact Analytics

About

AI Engineer at Impact Analytics with 2+ years of experience building production AI systems, LLM applications, and data-driven machine learning solutions that solve real business problems. I specialize in developing end-to-end AI and ML systems, including data pipelines, feature engineering, model development, evaluation, and deployment. My work involves building scalable AI applications using Python, SQL, FastAPI, Redis, and Google Cloud Platform, integrating LLMs, LangGraph orchestration, RAG pipelines, and vector embeddings to enable intelligent analytical systems. At Impact Analytics, I have designed systems including Text-to-SQL engines, agentic AI assistants, causal inference models using DoWhy, and computer vision pipelines using CLIP to automate analytics and business decision workflows. My technical interests include Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Deep Learning, NLP, Computer Vision, Causal AI, and AI Agents, with experience in frameworks such as LangGraph, LangChain, Pytorch, and LlamaIndex. With a Strong Foundation at NIT Warangal and Hands-on Industry Experience , I am excited to leverage these skills to create a real impact. 📩 Email: joy.gangopadhyay9@gmail.com 📞 Phone: +91 8758179671

Location

Bangalore Urban

AI stack

antigravity

opencode

Work experience

AI Engineer at Impact Analytics, Jun 2024 - Present · 2 yrs 3 mos, Bengaluru, Karnataka, India, Architected a production-grade Agentic AI BI platform using LangGraph, LangChain, Redis memory, and HITL validation workflows, enabling stateful multi-agent orchestration, contextual reasoning, and high-accuracy enterprise decision support. Engineered an LLM-powered intelligent query routing framework with dynamic intent classification across Insight, Causal, and Data agents, alongside a Text-to-SQL pipeline for autonomous NL-to-query generation and analytical response synthesis. Developed an end-to-end causal inference ML engine using DoWhy, GCMs, and KPI-level DAG modeling to perform root-cause analysis, counterfactual reasoning, and explainable AI-driven business impact assessment. Built a taxonomy-aware semantic NER and entity resolution pipeline leveraging vector embeddings, similarity search, and contextual retrieval to improve intent disambiguation, query understanding, and analytical precision. Implemented scalable RAG and observability pipelines with semantic retrieval, metadata filtering, Redis caching, async execution, prompt tracing, and hallucination mitigation techniques to optimize inference latency, grounding, and production reliability.

Education

Bachelor of Technology - BTech, National Institute of Technology warangal

Senior Secondary Board, Science, Lal Bahadur Shastri Group of Institutions

Public links

github: https://github.com/joy206122