Pallab Roy - AI Engineer
I engineer production-grade, zero-data-retention AI pipelines—utilizing AI agents, Machine Learning, LangChain, LangGraph, and RAG—that bridge the gap between complex raw data and deterministic business automation.
Currently, I am partnering with founders of high-compliance advisory firms to audit and automate their internal workflows, leveraging half a decade of professional AI engineering and full-stack development experience to deliver production-ready systems.
At BASSETTI India I shipped:
- Multi-agent compliance screening engine: Designed to automate federal grant (FEMA) RFI reviews utilizing an Orchestrator-Worker architecture.
- Zero-hallucination AI parsing engines: Architected to eliminate manual data entry bottlenecks, structuring chaotic field data into strict, compliant database formats, decreasing manual processing time by 95%.
Outside of work:
I am incredibly proud of designing robust, complex state-testing architectures and authoring trench-tested engineering deep-dives. I focus on building systems that process high volumes—supporting user bases of ~7000 or processing ~1000 daily API requests with high availability across containerized deployments.
I write about Agentic AI design and production MLOps engineering on Dev.to and LinkedIn.
I am actively pursuing remote software engineering and AI architecture roles within the US, European, and Singaporean tech hubs that prioritize building secure, human-in-the-loop ML applications.
If you are a tech-forward firm looking to reclaim senior billable hours, or an engineering team hiring a pragmatic, AI-native developer, send me a DM on LinkedIn to talk architecture.