Announcing the Kilo League
A Yearly Competition Highlighting the Best in Agentic Engineering
TL;DR: Kilo is hosting weekly, virtual challenges throughout the year that will culminate in a grand prize of $50,000.
The first challenge, where top prize gets $500 in Kilo Credits, is live here:
We are thrilled to announce the launch of Kilo League—a year-long competition to push the boundaries of what's possible with AI coding agents.
Remember when hackathons weren’t all the same?
And remember when you felt that rush of using technology that is truly new and ahead of the curve?
AI coding tools are not just simple assistants. We believe that many developers have barely tapped the surface of what’s possible with AI coding agents. The Kilo League is designed to help you become a natural at Agentic Engineering—shifting your role from writing syntax to conducting intelligent agents.
What is the Kilo League?
Kilo League is the world’s first competitive arena for the full AI engineering lifecycle. Building on the success of our App Builder challenge, we’re launching an entire league to keep you moving.
Kilo League isn’t just a hackathon; it is a season-long journey of competitions, learning tracks, and community to help you master the art of:
LLM-Driven Architecture: Going from a vague prompt to a robust system design.
Agentic Coding & Debugging: Using AI to write, refactor, and fix code faster than ever before.
Deployment & Optimization: Pushing to production and tuning performance with autonomous agents.
Multi-Agent Workflows: Orchestrating teams of agents to handle complex, parallel tasks.
Starting this week, we are opening the arena to everyone—from students to senior architects—to compete for a grand prize package worth $50,000 and over $1,000 in weekly credits and rewards.
The Goal: Mastery
Most competitions measure how fast you can type. The Kilo League measures how well you can orchestrate.
Your Kilo Account: Your progress is tracked via your unique Kilo Account. Every challenge you complete builds your “Agentic Profile,” showcasing your ability to work with AI agents effectively.
Weekly Challenges: At least twice a month, we’ll be dropping a new “Speedrun.” These aren’t just coding problems; they are engineering scenarios. You might be asked to architect a database schema using only natural language, or deploy a serverless app using a multi-agent swarm.
Quarterly Majors: Top earners will be invited to exclusive virtual hackathons to qualify for the end-of-year Championship bracket.
We’re partnering with top frontier models, open-source heavyweights, inference providers and more. As a League member, you’ll get early access to the best tech out there—and can earn a chance to put your work in front of the best engineering teams in the world.
Kilo League Challenge #1: Automate Everything with Cloud Agents + Webhooks
Kilo’s Cloud Agents run in the cloud, which means they can be triggered programmatically via webhooks.
This opens up a whole world of automation possibilities: CI/CD pipelines that spin up agents on PR events, integrated bots that kick off coding tasks, scheduled jobs that run maintenance scripts - you name it.
The Challenge:
Build something cool that uses Cloud Agents + Webhooks to automate a workflow.
Could be a GitHub integration, a Discord bot, a cron job that refactors code every night at 3am... get creative!
Check out this blog post for inspiration and examples of what’s possible.
Ready to Join?
The arena is open. All you need is a Kilo account and the drive to compete.
Create a Kilo account.
Enter your first Kilo League challenge here
Join the Discord (optional, but recommended!): Connect with other Agentic Engineers and get live updates on the latest drops.
Welcome to the Kilo League. See you in the arena!





LFG!!
This is brilliant timing tbh. The shift from viewing AI as copilot to orchestrator is smething most devs are still figuring out. I've experimented with multi-agent setups for prototyping, and honestly the hardest part wasn't the code but learning how to delegate properly. Competitons with real-world scenarios acelerate that learning way faster than reading docs.