Professional Overview
I build production Generative AI systems that let non-technical users query enterprise data in plain language. My work centers on Agentic RAG and Text-to-SQL platforms using Python, LangChain, LangGraph, FastAPI, PostgreSQL, vector databases and AWS. At IBM, I designed CAQSLO, a dual-agent system (Task Planner and Task Executor) that lets logistics teams query shipment, SLA, contract and delivery-proof data across SQL, documents and images through MCP adapters. It replaced manual cross-referencing of separate data sources with a single natural-language interface, backed by LangSmith tracing, Mem0 memory and output guardrails.
Experience
AI Expertise
Skills
Employment Type
Description of Responsibilities
- Built dual-agent AI system for natural-language querying of logistics data
- Built async agent workflows with FastAPI, arq and Redis
- Deployed on AWS (EKS, RDS, S3) using Terraform
- Added LangSmith tracing, Mem0 memory and output guardrails
Key Achievements / KPIs
- Enabled non-technical users to query 4 business data domains in plain language
- Unified 3 data modalities (SQL, documents, images) in one agent workflow
Education
Field of Study: Biotechnology
Areas of Expertise
AI & Machine Learning Engineers
Design, train, and ship production ML models — from data pipelines to model inference at scale. Covers supervised learning, neural networks, LLM integration, and model optimisation.
Data Science & Analytics
Extract business insights from raw data using statistical modelling, predictive analytics, A/B testing, and interactive dashboards.
MLOps & AI Infrastructure
Architect scalable ML infrastructure — CI/CD for ML, model versioning, experiment tracking, monitoring, and cloud deployment on AWS, GCP, or Azure.
- Open to
- Full-time
- Time zone
- Kolkata (UTC+5.5)
- Work language
- English