At Infosys, I engineer autonomous multi-agent workflows with LangGraph and LangChain to automate complex, multi-step enterprise business processes. I also work on runtime scheduling and state memory.
At Infosys, I develop Python and Node.js REST APIs with Pydantic validation and OAuth/IAM access controls. I implement Agent-to-Agent communication and MCP to connect agents with enterprise systems and third-party SaaS platforms.
I use Langfuse and LangSmith to trace execution and address workflow bottlenecks, and I build evaluation pipelines in Galileo to validate LLM outputs. I also containerize AI applications with Docker and deploy them on Kubernetes through CI/CD pipelines.
On my AI Invoice Auditor project, I created a multi-agent document extraction and schema validation pipeline, with AWS infrastructure provisioned using Terraform. I also implemented RAG with FAISS; in a separate churn prediction project, I developed and benchmarked machine-learning pipelines that achieved 92.46% accuracy.

