AI Harness: From Scratch to Product (Students only)
(Students only event)
AI is moving beyond simple chatbots and prompt-based applications.
The next step is building AI systems that can reason, use tools, maintain context, and actually perform tasks.
AI Harness: From Scratch to Product is a hands-on technical session by CBE Devs designed to take you through the journey of building an AI harness from the ground up — from understanding the architecture and connecting an LLM to giving it tools, memory, and an agent loop.
This isn't a session about simply using an AI tool.
We'll build one.
We'll start with the foundations, work through the core components, and bring everything together into a working AI harness that can be tested and demonstrated live.
What we'll get into
The session will be practical and implementation-focused. Expect to explore:
Understanding AI Harness architecture and setting up the development environment
Integrating an LLM into an application
Giving AI access to tools and function calling
Building the agent loop that allows an AI system to reason and act
Understanding and implementing memory and context
Putting everything together to build and test a working AI harness
Build it. Understand it. Run it.
What we'll build
Throughout the session, we'll look at practical examples of what these building blocks can enable:
AI Research Assistant
Tool-Using AI Agent
Autonomous AI Harness
The goal is not just to watch a demo, but to understand what's happening underneath and learn how these systems can be built from scratch.
Who should come?
This session is for students who want to move beyond using AI applications and understand how modern AI-powered products are actually built.
You don't need to be an AI expert.
A basic understanding of Python and a willingness to build is enough.
What you'll need
Please bring your own laptop.
Prerequisites
Basic Python knowledge
Python 3.10+
VS Code
Facilitated by
Jayaragul N & Jayasurya M
CBE Devs
This is a hands-on session focused on learning by building.
Come with your laptop, curiosity, and a willingness to experiment.
Build it. Understand it. Run it.