Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Raptor is presented by its creators as an open-source, self-hostable platform for hackathon submissions and judging. Its distinctive idea is to make key decisions—submission eligibility, score calculation, reviewer access and administrative actions—visible and reviewable rather than asking participants to accept an unexplained result. Those are project-author descriptions, not independently verified software or security findings.
What Raptor is designed to do
Raptor is described as a system for collecting hackathon projects, assigning judges, applying a scoring rubric and publishing outcomes. Its repository calls it a “self-hostable, open-source hackathon submission & judging platform” built “backend-first.” The project article frames the goal as making consequential decisions checkable: “If a platform decides who wins money and recognition, every step of that decision should be something you can actually check, not just something you have to trust.” The Raptor project article and the repository description are the sources for these claims.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Hackathon Mode: On - Programmer Hackathon Design T-Shirt | $21.99 | Buy on Amazon |
How Raptor checks hackathon submissions
According to the project authors, Raptor checks submitted projects against GitHub commit history and the event window. A repository that is private, is not hosted on GitHub, or has commits outside that window is flagged for organizer review rather than automatically treated as eligible. The described workflow also requires a permanent written reason for a disqualification. This distinction matters: an automated flag is a prompt for human review, not by itself a final ruling.
Recommended Free Tools
The unresolved question of which code is judged
Eligibility is only part of submission integrity. A repository branch can move after a submission is accepted, potentially leaving uncertainty about which version judges actually evaluated. In an October 1, 2026 comment, a Raptor author said a fix had been tested locally so judges would resolve a pinned GitHub commit. The same comment said GitLab and Bitbucket would use organizer-confirmed binding and that the change had not yet been pushed. Therefore, the available project account does not establish that commit pinning was released or deployed; organizers should verify the current repository and running version before relying on it.
#1 Best Overall
- Programming Hackathon design. Hackathon Mode: On
- A must-have for any coder, software developer, or hacker. Perfect for the next hack day or hackathon event.
- Lightweight, Classic fit, Double-needle sleeve and bottom hem
How judging scores are calculated
The authors describe judges receiving assigned projects and scoring them against a rubric, with bonus tracks and written feedback. A judge may revise a review until judging closes. The article’s illustrative scoring chain is:
- Apply the rubric’s weights to the criterion scores.
- Add any applicable bonus points after the weighted rubric component.
- Average the reviews submitted by judges who completed them.
- Scale the result to the event’s display scale.
This is a description of the project’s example, not a universal formula or a reported performance benchmark. The article does not establish how a specific event configures its rubric, bonus tracks, display scale or handling of missing reviews beyond the stated use of completed reviews.
What judge-level normalization is meant to address
Judges can use scoring scales differently: one may score generously while another reserves high marks for only a few projects. The authors describe optional normalization based on each judge’s scoring profile, intended to reduce the influence of those differences. They report a simulation involving 30 judges and 40 projects, saying normalization helped more when judges evaluated many projects and had a smaller effect when they evaluated few. The cited account does not provide enough underlying data or methodology to reproduce the simulation, and it reports no improvement percentage. Treat normalization as an optional approach described by the authors, not proof that rankings become more accurate or fair.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchRoles, permissions and audit records
The project article describes separate capabilities for visitors, participants, judges, organizers and admins. Judges see assigned work; organizers manage event operations and publication; and admin-level access is described as logged. The goal is to keep judging and event administration distinct while preserving a record of privileged actions. These are feature claims by the project authors, not an independent security audit. A real deployment still needs organizers to inspect the implementation, configuration and logs for their own requirements.
How the self-hosted setup is described
The reported architecture uses a Next.js web application, NestJS API, BullMQ worker, PostgreSQL and Redis. The authors describe five Docker Compose services split across application and data networks, with the database and Redis not exposed and the browser communicating with the API rather than directly with the database. They say that docker compose up is sufficient to run the platform and that no cloud account, external API call or signup is required. These statements describe the project article’s setup; they do not verify that every current release or deployment has the same requirements.
What Dogfood 2026 adds—and what it does not prove
Hackathon Raptors’ separate Dogfood 2026 brief describes an event challenge to build a submission and judging platform for real self-hosted use. Its stated priorities include backend-enforced role isolation, weighted rubrics, normalization, audit trails, protections against voting abuse, documentation and local operation. The brief weights its own judging categories as follows:
| Dogfood 2026 category | Weight in the event brief |
|---|---|
| Tier completion and correctness | 40% |
| Judging integrity | 25% |
| Adoptability and operability | 20% |
| Code quality and innovation | 15% |
These are Dogfood’s criteria, not Raptor’s judging weights. The event brief says Dogfood ran for 72 hours, offered a $2,500 prize pool and was scheduled for September 25–28, 2026, with judging through October 8 and winners announced October 9. It also claims Hackathon Raptors had run 35 hackathons across 85+ countries since 2023; those are organizer-published figures, not independently corroborated here. For changes to event details, the brief directs readers to the official event website and Discord as the source of truth. Read the Dogfood 2026 event brief.
What organizers should verify before using Raptor
The project’s stated design addresses real trust questions, but a description of intended behavior is not evidence that a particular deployment enforces it correctly. Before an event, organizers can turn the project’s claims into checks tied to their own rules:
Quick Recap
- Confirm the installed version and whether accepted submissions are bound to an immutable commit or equivalent evidence.
- Test how out-of-window commits, private repositories and non-GitHub links reach organizer review, and how a disqualification reason is retained.
- Inspect rubric weights, bonus handling, incomplete reviews and the conversion to the event’s displayed scale.
- Decide whether normalization fits the judging design, and document its effect for participants and judges.
- Verify permissions and audit records using the actual deployed configuration, especially for administrative access.
- Check that the Compose deployment’s network exposure and data handling match the event’s operational and security needs.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
