About me
I'm Shashank Sahu — an AI Engineer who builds autonomous systems that actually work in production. My focus is on agentic AI: multi-agent orchestration, RAG pipelines, LLM infrastructure, and the hard engineering problems that come after the model works — observability, governance, memory, and reliability at scale. I work primarily with LangChain, LangGraph, Python, and cloud-native tools across AWS and Azure.
I've shipped AI systems handling real financial workflows at Dailoqa, built evaluation frameworks for RAG pipelines from scratch, and integrated LLM-powered features into platforms serving tens of thousands of users. Before AI, I spent two years building production-grade full-stack systems — a festival portal processing 15,000+ registrations, high-traffic websites for ISKCON Ghaziabad, and a cross-platform mobile app at Peece Technologies. That foundation in shipping and operating at scale is what grounds my AI work today.
I'm not chasing hype. I care about building AI that is useful, auditable, and production-ready — systems that make real workflows genuinely better. I believe the best outcomes come from clear thinking, thoughtful collaboration, and staying relentlessly curious. I'm open to roles and collaborations in AI engineering, agentic systems, and LLM infrastructure.
What i'm doing
-
Agentic AI Systems
Designing and building multi-agent pipelines using LangChain, LangGraph, and AutoGen — autonomous systems that reason, plan, use tools, and operate reliably in production.
-
RAG & LLM Infrastructure
Building retrieval-augmented generation pipelines, evaluation frameworks, and LLM-powered features with observability, guardrails, and governance baked in.
-
Backend & API Engineering
Building fast, scalable backend services and APIs using Python, FastAPI, Node.js, and cloud infrastructure across AWS and Azure.
-
Full-Stack & App Development
Shipping production-grade web and mobile applications using React, Next.js, and Flutter — from high-traffic portals to cross-platform mobile apps.
Testimonials