Deepak Agrawal

About Deepak Agrawal

I am a Principal Engineer and Architect with over 20 years of experience building massive-scale distributed systems and enterprise products. My background is rooted in \"solid engineering\"—developing core server-side systems using Java and Python—but my career has evolved into architecting cloud-native AI platforms on OCI and AWS.Currently, I am focused on the intersection of Agentic AI and Modern Data Architectures. I don’t just integrate APIs; I design autonomous Agentic Services that reason, plan, and execute workflows, backed by robust data lakehouses.My Core Focus: CLOUD AI PLATFORMS (OCI AI STUDIO & FORECASTING) During my tenure as a Cloud Architect at Oracle, I helped build AI Studio (Health & AI Data), a privacy-focused platform for managing datasets and custom annotations for sensitive industries. I also architected the OCI Time-Series Forecasting Service from scratch, designing it to handle minute-level granularity for high-scale prediction workloads.🤖 AGENTIC AI & GEN AI I am actively building systems using CrewAI, LangChain, and LangGraph. My work spans the full model lifecycle—from training and fine-tuning to rigorous model evaluation—ensuring LLMs are reliable enough for production. DATA ENGINEERING & LAKEHOUSES AI requires clean, unified data. At Cloudera, I am designing Apache Iceberg REST catalogs and Cloudera Metalake to stream metadata discovery across heterogeneous environments. I specialize in unifying data silos (Hive, HMS) to enable seamless analytics.I am passionate about solving the \"hard problems\" where Data Engineering meets Artificial Intelligence, transforming raw data into actionable, autonomous intelligence.

Experience

1. Principal Engineer

   Cloudera

   Feb 2024 — Present · Bengaluru, IN

   Cloudera Metalake: Cloudera Metalake is a centralised federatedService designed to streamline metadata discovery, access, and governance across multiple heterogeneous Iceberg REST catalogs. It addresses the inefficiencies of managing multiple independent data catalogs by providing a unified endpoint, reducing operational complexity, and enhancing data security and governance.2. Rest Catalog (Data sharing): Rest catalog is an Iceberg standards unify open data service that allowsorganisation to access structured or unstructured data from CDP catalog with zero data copy & movement.

Education

• Madhav Institute of technology & Science Gwalior

  Bachelor of Engineering (B.E.), Computer Science

  2002 — 2006

• IIT Bombay

  e-Postgraduate Diploma (ePGD) in Artificial Intelligence and Data Science

  2026 — 2027

Skills

• Odata
• Microsoft Sql Server
• Informix 4gl
• Spring Framework
• Tfs
• Microsoft Visual Studio C++
• Java
• Oracle 9i
• C++
• Html
• Struts
• Kerberos
• Sap Hana Vora
• Apache
• Soa
• Servlets
• Unix Shell Scripting
• Maven
• J2ee Application Development
• Node.js
• Mysql
• Jdbc
• Xml
• Informix
• Perl
• Sap Lumira
• Shell Scripting
• Red Hat Linux
• Sap Hana
• Spring
• Ajax
• Sso
• Informix Sql
• Jsp Development
• Microservices
• Clearcase
• Javascript
• Web Services
• Clearquest
• Unix