Agile Pune 14-15 Nov 2025 | From Pilot to Platform: How Agentic AI Transforms Platform Engineering by Supriya Rao

Agile Pune 14-15 Nov 2025 | From Pilot to Platform: How Agentic AI Transforms Platform Engineering by Supriya Rao

• 1.
  Disclaimer by Speaker •Views, thoughts, and opinions expressed in the session and presentation, collectively referred to as "the content", belong solely to me in my personal capacity, and not necessarily to my employer / organization / client, etc. • "The content" is based on my learning, experience as well as knowledge gathered through material available in the public domain. • I do not endorse or promote any organization, committee, product or person through this session. • I have agreed to the Code of Conduct, Privacy Policy, Speaker Engagement Policy as referenced in the Call for Proposal submitted by me or on my behalf on the Agile Pune Conference website (www.agilepune.com). Supriya Rao 14th Nov 2025
• 2.
  PRACTICAL LESSONS FROM BUILDINGAN AGENTIC-FIRST DEVELOPER PLATFORM FROM HYPE TO AN AUTONOMOUS ROUTE TO LIVE Supriya Rao, MD India
• 3.
  ACT 1 THE LANDSCAPE HYPE,REALITY, & THE EVOLVING PLATFORM MANDATE
• 4.
  AGENTIC AI ISSUDDENLY EVERYWHERE. HERE’S HOW COMPANIES ARE EVALUATING AND USING THESE BUZZY TECH TOOLS FORTUNE THE NEXT “NEXT BIG THING”: AGENTIC AI’S OPPORTUNITIES AND RISKS UC BERKELEY “AGENTS ARE NOT ONLY GOING TO CHANGE HOW EVERYONE INTERACTS WITH COMPUTERS. THEY’RE ALSO GOING TO UPEND THE SOFTWARE INDUSTRY, BRINGING ABOUT THE BIGGEST REVOLUTION IN COMPUTING SINCE WE WENT FROM TYPING COMMANDS TO TAPPING ON ICONS.” Bill Gates THE THIRD WAVE OF AI IS HERE: WHY AGENTIC AI WILL TRANSFORM THE WAY WE WORK FORBES AGENTIC AI: 4 REASONS WHY IT’S THE NEXT BIG THING IN AI RESEARCH IBM HYPE, REALITY, AND THE ENTERPRISE CHASM
• 5.
  ...AND THE ENTERPRISEREALITY
• 6.
  THE EVOLVING MANDATE POWERINGA COLLABORATIVE ECOSYSTEM DEVELOPER PLATFORM SELF SERVICE PLATFORM APIS
• 7.
  DATA INTEGRATION TRUST THEGAPS ARE BARRIERS TO COLLABORATION
• 8.
  ACT 2 THE MINDSET &THE COMPASS FINDING YOUR FIRST COLLABORATIVE USE CASE
• 9.
  PATH 2 THE STRATEGICCHOICE ONE-OFF TOOLS VS. A FOUNDRY MINDSET ONE OFF TOOLS A FOUNDRY MINDSET FOUNDRY PLATFORM PATH 1
• 10.
  INTRODUCING THE DIGITALFOUNDRY & ITS PRIMITIVES MODEL GARDEN AGENT FRAMEWORK RUNTIME AND SERVICE LAYER OBSERVABILITY SUITE
• 11.
  GENERATING TEST SCAFFOLDING (5) PLOTTING YOURROUTE ENGINEERING USE CASES TECHNICAL FEASIBILITY AUTO FILL PR DESCRIPTIONS FROM GIT DIFF (2) AUTONOMOUSLY FIXING NOVEL SECURITY VULNERABILITIES (?) SHIFT LEFT ANALYSER (4) SUMMARISING INCIDENT POST-MORTEMS (6+) ENRICHING THE PR PROCESS (2) GOLDILOCKS ZONE ENGINEERING VALUE NEW API PLACEHOLDER ANALYSER (5) TEST FAILURE EXPLAINER (2) SECURITY THREAD MODELLING (?) Internal SDK USAGE AND DEPRECATION (3) RESEARCH AND TECH RADAR (3) TECH DEBT TAGGER (5) BACKLOG ASSISTANT (3) # = tool count INTEGRATION COMPLEXITY DISTINCT SYSTEMS, DATA SOURCES, APIS
• 12.
  ACT 3 THE EVOLUTIONOF A USE CASE THE PR AGENT
• 13.
  THE PR AGENT(V1.5) DEEPER CODE ANALYSIS PR ENRICHMENT AGENT GITHUB REPO V2 GITHUB CONNECTOR V1 MODEL GARDEN V2 AGENT FRAMEWORK PRIMITIVES COMPLEX MULTI-TURN ITERATION ON THE CODEBASE
• 14.
  THE PR AGENT(V2) IS NOW JIRA-AWARE PR ENRICHMENT AGENT V2 GITHUB CONNECTOR V1 MODEL GARDEN V2 AGENT FRAMEWORK V1 JIRA CONNECTOR GITHUB REPO JIRA PRIMITIVES
• 15.
  ACT 4 PREPARING THE GROUND THEENTERPRISE READINESS CHECKLIST
• 16.
  1. HOW DOOUR AGENTS SAFELY ACCESS SYSTEMS? 2. HOW DO WE CONTROL OUR DATA AND MODELS? 3. WHERE DO THESE AGENTS ACTUALLY RUN? THE THREE ENTERPRISE QUESTIONS WE MUST ANSWER
• 17.
  BLOCKER 1 SYSTEM ACCESS THEAPI GATEWAY PROBLEM AGENTS ARE USELESS IF THEY CAN'T INTERACT WITH OUR ENTERPRISE SYSTEMS. SOLUTION We mandate an API-First culture and create agent-specific service accounts with least-privilege access. Our APIs are the standardised, safe docking ports on the foundry's assembly line.
• 18.
  BLOCKER 2 DATA &MODEL GOVERNANCE THE PRIVATE FURNACE PROBLEM WHERE IS OUR PROPRIETARY DATA GOING? SOLUTION: We use Enterprise Model Gardens (Vertex AI, Bedrock, etc.) to keep data private. This is our private, certified furnace, and we build our own Access Control Layer on top to act as its control panel.
• 19.
  BLOCKER 3 THE DEPLOYMENTENVIRONMENT THE CLEAN ROOM PROBLEM WHERE DOES AN AGENT LIVE AND RUN? SOLUTION We treat agent hosting as a first-class platform product, deploying agents into Insulated, Sandboxed Runtimes—the foundry's 'clean room'. This secure, monitored environment, managed by a dedicated platform team, ensures both safety and velocity, enabling our platform to grow organically.
• 20.
  ACT 5 SCALING THE FOUNDRY THEFLYWHEEL EFFECT
• 21.
  THE FLYWHEEL EFFECT OURNEXT USE CASE IS NOW EASIER INCIDENT RESPONSE AGENT V3 GITHUB CONNECTOR V2 MODEL GARDEN V3 AGENT FRAMEWORK V2 JIRA CONNECTOR SANDBOX V1 AGENT RUNTIME GITHUB REPO JIRA SANDBOX PRIMITIVES
• 22.
  CONNECT WITH US LINKEDIN.COM/IN/SUPRIYA-Y-RAO/ supriya.rao@clearroute.io https://clearroute.io