Live
AI Agent Hackathon
Deadline Jul 3, 11:59 PM PT
Harbor x NovitaAgent Sandbox Hackathon
Run TB2.1 agents in Novita Agent Sandbox, publish on Harbor Hub, and compete for Novita credits.
Prize and credit eligibility: submit your public Harbor Hub result in Discord after uploading your job.
4Tracks
89TB2.1 tasks
$120Starter credits
$700Track prize
Live leaderboard
Public jobs only
| Rank | Track | Agent | Harbor Job | Model | Environment | Reward | Duration |
|---|
Public track datasets
| Track | Dataset | Tasks | Categories | Best for |
|---|---|---|---|---|
| Code & Debug | Dataset | 32 | software-engineering + debugging | Coding agents and repository work |
| Systems & Security | Dataset | 22 | system-administration + security + mathematics | Shell, security, and crypto-style math |
| Data & Science | Dataset | 26 | data-science + scientific-computing + ML + data-querying | Analysis, ML, and query tasks |
| File & Recovery | Dataset | 9 | file-operations | Filesystem recovery and transformation |
| Overall | Official TB2.1 leaderboard | 89 | All original categories | Official |
Startup Program
Building agents, developer tools, evals, or sandbox-heavy products? Eligible AI-native startups can apply for up to $10,000 in Novita cloud credits.