Shubhankar M.

About Me

S.M. is a Generative AI Engineer with 4 years of experience building production-grade LLM and multi-agent systems. He has hands-on experience in LLM fine-tuning, Retrieval-Augmented Generation, context and memory management, and AI observability frameworks. S.M. has a proven ability to translate business requirements into robust, scalable AI workflows integrated with Snowflake for data-aware reasoning and decision automation. He has experience with Publicis ReSources and HCLTech.

AI, ML & LLM

FAISS Langsmith Langgraph Machine Learning

Backend

Database

Vector Databases

QA & Testing

Other

Excel Macros Amazon S3 Beautifulsoup LoRa Prompt Engineering Agentic Al Generative Al

Work history

Publicis ReSources
Agentic Al Developer-Engineer
2025 - 2026 (1 year)
Remote
  • Designed and deployed agentic AI systems using LangGraph for multi-agent workflows.

  • Built Python + FastAPI+Prompt Engineering-based microservices to serve AI agents.

  • Implemented RAG pipelines using Vector Databases (FAISS / Snowflake Cortex Search) for semantic retrieval and grounding.

Langgraph PythonFastAPIPrompt Engineering Vector Databases FAISS Snowflake Cortex Search LoRa Langsmith TruLens
HCLTech
AI/ML Engineer
2021 - 2024 (3 years)
Remote
  • Automated 63+ Excel- Finance based reports reducing manual effort by 32 hours per cycle (3 FTE savings).

  • Designed ETL pipelines using Python Automation (Selenium, BeautifulSoup), stored data in Snowflake & Amazon S3.

  • Used Power Automate and Excel Macros for automated report scheduling, boosting consistency & timeliness.

SR Petrochem
Operation Manager
2018 - 2021 (3 years)
Remote
  • Overseeing end-to-end company operations—from procurement to distribution

  • Managing productivity, workforce, and reporting

  • Developing materials for sales, quality, and client engagement.

Education

Education
Post GraduateProgram inData Science andEngineering - Online
GREAT LAKES EXECUTIVE
2024 - 2024
Education
B.Tech
Jaypee University of Engineering and Technology
2014 - 2018 (4 years)