Hackathons

Use Lightning AI to organize hackathons to enable teams to work on ML projects together, on the cloud, with a set budget of cloud credits. Hackathon participants can publish their ML projects for anyone to run, or easily turn that into a viable business. This guide describes how to organize a hackathon with Lightning AI.

Set up a hackathon

Create an organization

First, create an organization with the hackathon name.

Then, invite participants to the organization.

Create a teamspace per team

Within the organization, create a teamspace per project or team. A teamspace groups studios under a single shared filesystem and grants access to those resources only to those team members.

Once the teamspace is created, assign the teamspace the following:

• credits
• access to the datasets they need
• access to specialized compute resources (as needed)

Publish seed templates

For teams to start building, we recommend you publish a few starter templates internally that are only visible to those organization members.

Organization members can duplicate and start from those templates.

One Studio per task

Within a teamspace, we recommend you structure each part of a pipeline or task as its own Studio. For example, create one Studio that trains a model and create a second Studio that deploys the model.

Publish the finished projects

Once the hackathon is done, teams can keep their Studios private or share them with the community for others to duplicate and run.

Advantages

Running hackathons on Lightning AI offers many advantages.

Set a budget

Lightning AI is the only platform in the world that lets you set a budget for a team to spend on cloud compute. This allows hackathon organizers to more efficiently use cloud credits and build more meaningful projects.

Cloud GPUs with zero setup

Most AI today requires cloud GPUs or access to high-performance compute resources. Hackathon participants can instantly start a Studio and get on cloud GPUs without any friction. The work they do there can extend to a real product.

From POC to production

Studios are used to deploy and scale some of the world's largest models across universities and enterprises. Hackathon participants can quickly turn a POC they built during a hackathon into a production system without having to set up everything again. Enable your hackathon participants to build real products and businesses faster than ever.

Organize teams

Hackathon administrators can create a teamspace for each different team so that they have a budget and isolate private data access.

Publish finished work

Studios can be made public and published to the world immediately after the hackathon finishes. Share on Twitter, social media, and instantly get prototypes and apps into the hands of millions of users. Participants can also deploy AI apps via the platform to get initial product-market-fit and scale their products.