intelligent Engineering: Reimagining How Software Is Built, Operated, and Evolved with AI Sahaj Software

DEVDAY_

intelligent Engineering: Reimagining How Software Is Built, Operated, and Evolved with AI

Abhishek Kumar, Karun Japhet, Dhavala Sai Sri Mahita, Sara Paul, Denny P Sam, Harshad Nawathe, Vignesh S, Praveen Shivanagoudar, Samarth G R

1 November 2024

Marathahalli – Sarjapur Outer Ring Rd, Bellandur, Bengaluru, Karnataka 560103, India

View Location

About the Event

Your team has access to every AI tool out there, yet the way your team builds and maintains hasn’t changed by much. Today, you’re not alone, but can you say the same of the future? There are people already cracking the code on AI in software delivery, changing how they design, build, test, and ship. Devday puts you in the same room as these thinkers and tinkerers. Share the space and take part in the conversation.

Session 1: Engineer the Environment, Not Just the Prompt

Speaker: Karun Japhet

Phase 1 of AI-augmented software development was prompt engineering, phase 2 was context engineering, which is hitting its ceiling. Phase three needs to be harness engineering—designing the scaffolding and safety nets that determine what the model builds and whether it’s allowed to proceed. This talk examines what a production-grade AI harness looks like and leave you with a concrete framework for auditing your current harness, and knowing exactly what to build next.

Session 2: Production support using AI
Speaker: Abhishek Kumar

An AI agent that investigates production issues across infrastructure layers, finds root causes, suggests fixes, and learns from every investigation to make the next one faster.

Session 3: From Goldfish to Engineer: Designing an AI Assistant That Remembers
Speaker: Denny Sam

LLMs forget. Every conversation starts cold. Restating “I’m working on this story, here’s the repo layout, here’s what we tried yesterday, here’s how I love to work” is the real cost of using an AI assistant at work — and most of the productivity wins are in eliminating that cost. Eva is my personal AI assistant, built on Claude Code. It’s not autonomous. What it does is remember: how I work, what I’ve already decided, which tasks are open, which experiments failed, what feedback I’ve given before. I’ll walk through the scaffolding — a beads-based task tracker, a memory system that loads on demand, error and feedback registries, and hooks that block me when I get sloppy on my own rules. The talk is opinionated. The discipline is what makes it useful: an assistant that pushes back when I drift, that verifies sticky notes against actual state before acting, that flushes context to a story file before switching tasks. Less autonomy, more rigor. The boring path which helps you stay on track!

Session 4: Reimagining SDLC with Claude Code
Speakers: Sara Paul and Mahita Dhavala

What happens when AI moves beyond a coding assistant and becomes part of how your entire engineering team works? In this lightning talk, we share how we embedded AI across the SDLC—from grooming and planning to building, reviewing, and testing—transforming isolated experiments into a scalable, team-wide capability. Learn the practical patterns we used to codify workflows, strengthen guardrails as adoption grew, and ensure engineers stayed accountable for AI-assisted output.

Session 5: Autonomy Requires Engineering | A field report on moving from AI-assisted to agent-driven delivery
Speaker: Praveen S and Samarth G R

AI-assisted delivery is changing how we build across Sahaj — and as the AI gets faster, more humans spend their time hand-holding it through every step. But autonomy isn’t a property of the model; it’s a property of the system around it. You earn it. This talk is on the climb to agent-driven delivery: cloud agents that own plan → code → iterate, humans on the loop for intent, approval, and merge.

Session 6: The Practical Guide to Reverse Engineering XXL Codebases with Agentic AI
Speaker: Harshad Nawathe

If you thought ‘infinite context’ windows can effortlessly swallow your enterprise monolith, then think again. This session shatters the illusion of massive prompt windows and introduces a practical, distributed-systems approach to AI context management. Discover the exact framework used to orchestrate an army of specialized, tool-driven subagents that can successfully map out and reverse-engineer 10,000+ file XXL codebases without ever hitting the context wall.

Session 7: What is iE?
Speaker: Karun Japhet

AI-assisted software development is a broad space: vibe coding, spec-driven development, multi-agent workflows, and dozens more techniques. 2025 was all about using AI across the full SDLC, not just code generation but analysis, design, testing, and deployment. What’s in store for 2026 and beyond as the tools keep getting better?

Panel discussion: How the job of a software engineer is fundamentally changing with AI-engineering
Moderator: Vignesh S

Agentic coding tools promise dramatic productivity gains – but they come with real costs, a steep learning curve, and open questions about what “good engineering” even means when an AI can write most of the code.

This panel explores: How do you and your team demonstrate value in this new landscape? What engineering skills matter more now – and which ones matter less? And how do you build a culture that adopts AI tooling without losing the craft?

Important Registration Information

This is an invite-only event with limited seats.

All registrations will go through a screening process. If your registration is approved, you will receive a confirmation email.

Meet our Speaker(s)

Abhishek Kumar

Solution Consultant

Abhishek is a Solution Consultant at Sahaj, working on a large-scale AdTech platform spanning multiple regions. Recently, he's been using AI agents to investigate production issues across infrastructure, turning tribal knowledge into a system that learns from every investigation.

Karun Japhet

Solution Consultant

I’m Karun, an engineer and consultant at Sahaj Software. I work on large-scale systems, architecture, and AI in software development. Most of my time goes into helping teams design systems that hold up as they grow especially as AI becomes part of how we build.

Dhavala Sai Sri Mahita

Solution Consultant

Mahita is a Solution Consultant at Sahaj, passionate about agentic AI and the evolving role of AI in software development. Her work focuses on how AI coding assistants can improve developer productivity, and help build software more efficiently. She is also an active speaker, sharing insights on building effective agentic AI systems through talks and workshops.

Sara Paul

Solution Consultant

Sara has over a decade of experience in software development, engineering best practices and designing scalable distributed systems. She is currently actively adopting AI-assisted engineering practices with tools like Claude Code to streamline software development.

Denny P Sam

Solution Consultant

Denny has been building software, data pipelines, and other data products for customers for over 8 years. He is currently working s on an adtech platform to optimise out-of-home inventory selection for large scale customers.

Harshad Nawathe

Solution Consultant

Harshad has over 16 years of experience building scalable software, distributed systems, and data products. With a background deeply rooted in core software engineering and agile practices, he is currently helping large enterprises on AI enablement and legacy system modernization—leveraging agentic AI to safely navigate, decode, and transform complex codebases.

Vignesh S

Solution Consultant

Vignesh is a Solution Consultant at Sahaj with over 10 years of experience building products and solutions across various domains like fintech, customer support. Currently working on a legacy Ad-Tech platform spanning multiple regions, he is exploring and championing team adoption of AI augmented workflows across the SDLC.

Praveen Shivanagoudar

Solution Consultant

Praveen is a Solution Consultant at Sahaj with two decades of experience in the industry. He specializes in building core platforms, data systems, and practical solutions to complex problems. Recently, his work has focused on developing AI-driven software systems.

Samarth G R

Solution Consultant

Samarth is a Solution Consultant at Sahaj with nearly a decade of experience building distributed systems, platforms, and developer-facing engineering capabilities across multiple domains. He's currently focused on how AI is reshaping software engineering beyond code generation — designing and experimenting with agent-driven delivery workflows that reimagine how software is planned, built, reviewed, and shipped, and the engineering foundations required for safe, scalable autonomy. He enjoys simplifying complex problems, challenging assumptions through good questions, and helping teams adopt practices that improve both delivery and developer experience.