

About us
Welcome to the Global AI Coimbatore Meetup! This group is dedicated to exploring the latest trends and applications of Artificial Intelligence, including AI and Society, Artificial General Intelligence, Azure SQL Database, recommender systems, and Federated discovery of big data. Whether you're a beginner or an expert, join us to learn about Creating simple Windows Azure applications, Data Management Platforms, and Artificial Intelligence: A Modern Approach. Let's come together to discuss and collaborate on all things related to AI in Tamilnadu.
Upcoming events
1

Harness Engineering from First Principles — Software Hacktory: 0 → 1
iamneo, 1205/A, 3rd Floor, SPA SRR Towers, Avinashi Rd, Pappanaickenpalayam, Coimbatore, Tamil Nadu 641037, India, Pappanaickenpalayam, TN, IN>>
Registration link - https://luma.com/dffqyhoi?tk=dCJ9b8
>>
Most software teams today use AI to generate code.
But what happens when we go beyond code generation and build systems where AI agents can reliably help plan, build, test, ship, track, and improve entire software products?
Software Hacktory: 0 → 1 is an in-person AI engineering event in Coimbatore focused on what it actually takes to move from capable AI models and impressive demos to reliable, real-world software systems.
The event brings together two complementary perspectives on that journey: Harness Engineering and building AI products beyond the demo.## Harness Engineering from First Principles
Speaker: Rishub C R
LinkedIn: /rishub-c-r
We'll explore Harness Engineering — the engineering discipline around building the systems, workflows, tools, guardrails, context, and feedback loops that turn capable AI models into reliable software-building agents.
The session will walk through how a software factory can be designed using modern agent harnesses, and how an idea or engineering ticket can move through an agentic software development lifecycle — from planning and implementation to testing, review, delivery, and observability.
What we'll explore- Building a software factory and agent-ready software stack from first principles
- Designing agents that can work with real engineering workflows and tools such as Jira and Linear
- Structuring planning → implementation → testing → review → delivery across an agentic SDLC
- Giving agents the right tools, context, permissions, and guardrails to actually get work done
- Tracking what agents are doing through traces, tool calls, costs, failures, decisions, and outcomes
- Running the same software factory across everything from local Small Language Models (SLMs) to frontier models
- Understanding the practical trade-offs between local vs cloud, cost vs capability, speed vs reliability, and autonomy vs control
- Exploring where humans should remain in the loop — and where agents can safely operate independently
This is not a vibe-coding session.
We're not going to press a button and ask AI to magically build a startup.
Instead, we'll look at the engineering underneath reliable agentic software development: clear interfaces, good context, tools, evaluations, feedback loops, observability, and human control.
Expect a mix of first-principles explanations, architecture discussions, practical examples, and technical demonstrations.***
## From AI Agents to Real-World Products: What It Takes to Build Beyond the Demo
Speaker: Rajkamal Mountbatton
CEO & Co-Founder, O2VEND
LinkedIn: /rajkamalmountbatton- Building an AI agent that works in a demo is one thing.
- Turning it into a product that can survive real users, real workflows, business constraints, failures, changing requirements, and production environments is something completely different.
- As the creator of O2VEND and CEO & Co-Founder of Jeyan Technologies, Rajkamal brings a builder's perspective on what it actually takes to move AI systems beyond experimentation and into real-world products.
- The session will look at the challenges that emerge when AI moves from a prototype into production — and the engineering, product, and business thinking required to make that transition successfully.
Together, the two sessions connect both sides of the problem:
How do we engineer capable AI agents?
and
How do we turn those capabilities into products that actually work in the real world?***
By the end of the event, you'll have a clearer mental model of Harness Engineering, how modern coding agents actually work, and what it takes to turn AI agents and prototypes into reliable real-world products.
1 attendee
Past events
18


