Open to opportunities

Vishal Kumar

@vishalkumar33

Message

Agentic AI engineer building retrieval-augmented and multi-agent LLM systems for reliable, scalable automation.

India

Message

What I'm looking for

I’m looking to build production-grade agentic AI and RAG systems—owning end-to-end pipelines, evaluation, and reliability. I want a team that values measurable performance, clean integrations, and fast iteration on real user workflows.

I’m an Agentic AI engineer with over a year of experience building retrieval-augmented generation systems and multi-agent workflows. I focus on end-to-end LLM-integrated pipelines—especially RAG chatbots and document extraction agents—that prioritize reliability and scalability.

In my recent internship at Digivatelabs, I worked with Google Cloud DocumentAI and VertexAI, trained and customized DocumentAI pre-built processors for domain-specific field extraction, and validated endpoints using Postman. I also built an extraction agent with LangChain to detect document types and extract their contents.

I’ve delivered full RAG pipelines, from loading and chunking documents to embeddings and vector-store retrieval using LangChain. I trace and debug LLM calls with LangSmith and define benchmarks using AI system evaluation frameworks to assess performance, reliability, and accuracy across agentic workflows.

Experience

Work history, roles, and key accomplishments

Agentic AI Developer Intern

Digivatelabs

Jan 2026 - May 2026 (4 months)

Trained and customized Google Cloud DocumentAI pre-built processors to improve domain-specific field extraction, and tested endpoints using Postman. Built a LangChain-based extraction agent and a RAG chatbot end-to-end, using LangSmith tracing and benchmarks to evaluate agentic workflow performance and accuracy.

Python Lang Chain LangGraph Google Cloud Document AI RAG Langsmith Prompt Engineering fastAPI Postman

Data Analyst Intern

HCDS Tech

Apr 2025 - Sep 2025 (5 months)

Cleaned and analyzed datasets, producing reports with Excel and Power BI and summarizing findings for the team. Contributed to a RAG-based project by leveraging vector databases and prompt engineering for LLM API integration, and wrote PostgreSQL queries to pull and aggregate reporting data.