Sunil Jadhav

Data & AI Engineer with almost 5 years of experience designing and deploying production-grade Generative AI applications, agentic workflows, and ETL/ELT data pipelines using Python. Promoted for owning end-to-end delivery of multi-agent AI systems built with LangChain/LangGraph orchestration, Retrieval-Augmented Generation (RAG), prompt engineering, and Model Context Protocol (MCP) tool integrations — containerised and deployed on Kubernetes (GKE) with GitOps CI/CD (Flux). Collaborates closely with product, data science, and engineering teams across architecture, code review, deployment, and optimisation. 95% NL-to-SQL accuracy · 60% pipeline efficiency gain · 97.9% database uptime · 30% query latency reduction

Data Engineer L3 (Promoted) at Archents IT India Pvt. Ltd. (2026-04 – Present)

• Lead the design of stateful multi-agent workflows using LangGraph, implementing conditional branching, tool-use nodes, and human-in-the-loop checkpoints integrated with enterprise databases and APIs.
• Containerise and deploy agentic AI and RAG services on Google Kubernetes Engine (GKE), managing rollouts and environment consistency through a GitOps CI/CD pipeline built with Flux.
• Drive migration of legacy ETL pipelines to a scalable ELT architecture on Azure, reducing operational overhead across multi-million-record datasets.
• Define team-wide standards for RAG pipeline design, including chunking strategy, vector store selection, re-ranking, and hallucination mitigation through output grounding.
• Partner with product managers, data scientists, and engineering teams through architecture and code reviews to align AI-system design with business requirements; mentor junior engineers on LLM integration and prompt engineering.
• Architect and deliver production-grade AI platforms using LangChain, LangGraph, and MCP, integrating enterprise LLMs with internal systems while establishing reusable frameworks, observability, and security best practices for scalable AI deployments.

Data Engineer at Archents IT India Pvt. Ltd. (2021-12 – 2026-03)

• Engineered Azure Data Factory and SSIS pipelines processing millions of records daily, delivering a 60% efficiency gain through parallelisation and incremental load patterns.
• Architected SQL Server databases with advanced indexing, partitioning, and query-plan tuning, sustaining 97.9% uptime and 30% faster query performance under peak load.
• Implemented Elasticsearch hybrid dense and sparse retrieval across multi-million-row datasets, enabling sub-second semantic search beyond traditional SQL capabilities.
• Designed and maintained 20+ Power BI and SSRS dashboards with DAX, row-level security, and automated refresh pipelines, reducing stakeholder time-to-insight by 20%.
• Built scalable data ingestion pipelines with automated schema validation, data-quality checks, and incremental processing, improving downstream reporting reliability by 15%.
• Developed production-ready AI applications using LangChain, LangGraph, and enterprise LLMs, enabling natural language querying and intelligent workflow automation over structured business data.
• Designed and implemented Retrieval-Augmented Generation (RAG) systems with vector embeddings, semantic retrieval, metadata filtering, and prompt engineering, improving contextual accuracy while reducing hallucinations.
• Integrated AI agents with SQL Server, REST APIs, and enterprise services through modular tool-calling workflows, with monitoring, logging, and error handling for reliable production deployments.

Bachelor of Technology in Information Technology – PES Modern College of Engineering, Pune (2018 – 2021)

Diploma in Computer Engineering – Government Polytechnic Jintur, Maharashtra (2013 – 2018)