Looking for a job

Shubham Thakur

@shubhamthakur23

Message

Data Scientist building production LLM, RAG, and conversational AI systems that automate real workflows.

India

Message

What I'm looking for

I’m looking for a role where I can own production LLM/RAG systems, build conversational and automation workflows, and ship with strong engineering practices (FastAPI, Azure, Docker) to deliver measurable impact.

I’m a Data Scientist with 4+ years of experience building production LLM systems, RAG pipelines, and conversational AI. I’ve delivered measurable outcomes like 90% automated ticket resolution at scale and built end-to-end solutions that teams can run reliably in production.

At Providentia Technologies, I deployed a GPT-powered semantic product discovery chatbot serving 100+ users and handling 1,000+ daily queries. I also built and shipped a Mistral LLM + GraphQL support automation pipeline with a FastAPI-backed approach, plus a real-time multilingual voice AI agent for Hyderabad Police using VAPI + n8n and fully local setup.

Before that, at BlackCoffer, I automated ETL for Google Ads API + BigQuery and reduced manual reporting by 100%. I’ve also built and optimized backend systems as a Java Software Engineer—improving API/SQL latency for hospital operations and creating a BERT-powered report generator—so I bring strong engineering rigor to my AI work.

Experience

Work history, roles, and key accomplishments

Current

Founder & Solo Developer

Current

TamperTrail

Jan 2026 - Present (5 months)

Built a self-hosted tamper-evident audit logging system using SHA-256 hash chaining with FastAPI, PostgreSQL, and Docker, including multi-tenant design with RLS, immutability triggers, and WAL-based ingestion (<10ms). Achieved 200+ container pulls and designed the system for SOC 2, HIPAA, GDPR, and CERT-In compliance with single-command Docker deployment.

fastAPI PostgreSQL Docker SHA 256 Hash Chaining Audit Logs Compliance Engineering

Data Scientist

Providentia Technologies

Oct 2023 - Jan 2026 (2 years 3 months)

Deployed a GPT-powered product discovery chatbot (100+ users, 1,000+ daily queries) using a MiniLM-L6-v2 RAG pipeline with ChromaDB and a FastAPI backend on Azure Docker. Built a Mistral LLM + eDesk GraphQL system achieving 90% automated ticket resolution, and created a real-time multilingual voice AI agent for Hyderabad Police using VAPI + n8n with local AI4Bharat models.

Python fastAPI ChromaDB Azure PostgreSQL Docker Mistral AI GraphQL VAPI

Data Scientist

BlackCoffer

Jul 2022 - Mar 2023 (8 months)

Automated ETL workflows integrating Google Ads API with BigQuery, reducing manual reporting by 100%, and deployed scheduled pipelines on Heroku with monitoring. Led zero-downtime cross-database SQL migrations and improved healthcare graph database Cypher query performance via indexing and schema redesign.

SQL ETL BigQuery Heroku Data Migration Graph Databases Cypher Monitoring

Java Software Engineer

ICT Health Technology Services

Jan 2019 - Jan 2020 (1 year)

Built Spring-based modules for a hospital management system used by 50+ hospitals, improving API and SQL latency. Developed a BERT-powered report generator that reduced manual tasks by 60% and implemented structured SQL extraction to enhance accuracy.

Java Spring Framework SQL API Development Performance Optimization BERT NLP Data Extraction