Headline

Generative AI Engineer | LLM, RAG & AI Agents | Python • LangChain • LangGraph | Building Production AI Solutions

About

I’m a Generative AI Engineer & Data Scientist specializing in LLM-powered applications, Retrieval-Augmented Generation (RAG), and AI Agents — taking AI systems from proof-of-concept to production. I work extensively with Python, LangChain, LangGraph, Vector Databases, and cloud platforms (AWS/GCP) to build scalable, cost-efficient AI solutions that solve real business problems. What I do best: • Build RAG pipelines for document intelligence, OCR, and enterprise search • Design and deploy AI agents for workflow automation • Optimize LLM cost, latency, and accuracy in production environments • Deploy end-to-end systems using FastAPI, Docker, and cloud infrastructure Experience highlights: • Reduced information retrieval time by 60% using custom RAG pipelines • Cut manual processing effort by 10+ hours/week through AI automation • Improved LLM output accuracy by 25–60% via prompt engineering and evaluation Currently open to Generative AI Engineer / AI Engineer / Applied Data Scientist roles where I can build high-impact, production-grade AI systems. 📩 Reach me at: satyam2373@gmail.com 🌐 Portfolio: https://satyamjadhav.me

Location

Pune District

AI stack

aistudio

anthropic apis

antigravity

claude code

claude

codex

cursor

hermes agent

kimi

langchain

langgraph

langsmith

lovable

mcp

openai

openclaw

v0

claudecode

hermesagent

Base44

Bolt.new

ChatGPT

Exa

Kimi Code

Langflow

OpenAI Agents SDK

OpenAI Playground

OpenRouter

Pinecone

Qdrant

Wispr Flow

Work experience

Product Engineer at Aurenix AI, Feb 2026 - Present · 6 mos, Pune City

AI Solutions Engineer at BrightLabs.in, Jul 2025 - Feb 2026 · 8 mos, India, • Built LLM-powered RAG pipelines using LangChain, LangGraph, and Vector Databases (Milvus/ChromaDB) • Implemented AI agents with LangGraph for autonomous task completion and workflow automation. • Optimized LLM prompts using few-shot learning, chain-of-thought reasoning, and structured output formatting. • Reduced information retrieval time by 60% and manual effort by 10+ hours/week. • Deployed solutions using FastAPI, Docker, AWS.

Data science Intern at AdvaRisk, Jan 2025 - Jul 2025 · 7 mos, Pune, • Architected RAG pipeline, integrating Milvus vector database with Hybrid search for geographical data analysis • Spearheaded prompt engineering initiatives to optimize Large Language Model outputs for financial risk analysis. • Conducted A/B testing on GPT-4, Claude, and Gemini models to determine optimal cost-performance ratios. • Collaborated with the engineering team to clean and preprocess Large Amount of unstructured data, with hundred's of GB Data. • Evaluated LLM performance using RAGAS, BERTScore, and custom rubric-based metrics, reducing inference costs upto 40% through prompt compression, caching strategies, and model quantization • Improved model accuracy by 25% through rigorous testing and validation of AI-generated outputs.

AI R&D Engineer at HCLTech, Sep 2022 - Mar 2024 · 1 yr 7 mos, Pune, • Developed a prototype for automated mathematical rule extraction, converting unstructured PDF data into machine-readable expressions. • Engineered a End-to-End pipeline for Table Detection and Extraction from complex PDFs, achieving a 91% success rate, including unstructured data. • Conducted deep research on data extraction methods and presented findings to senior leadership to guide future R&D directions.

Education

Bachelor of Technology - BTech, Artificial Intelligence & Data Science, Vishwakarma Institute of information technology

HSC, Mathematics and Computer Science, Arihant college of arts commerce and science,Pune

Public links

github: https://github.com/satyam237