-
-
Home Page - Hero Section
-
Home Page - Info
-
Home Page - Principle
-
Home Page - FAQ
-
Report Page - Chat
-
Track Status Page - Report Status
-
Track Status Page - Incident Summary & Case Updates
-
Track Status Page - Communication Channel
-
Admin Portal - Track all Cases
-
Admin Portal - Detailed info for a particular Case
-
Admin Portal - Detailed info for a particular Case (Continue)
Inspiration
Beacon AI was inspired by one problem: people want to report wrongdoing but fear risks to their identity and career. We found that trust, anonymity, and evidence credibility are the main barriers, both in public systems and inside organizations, shaping Beacon AI’s focus on anonymous and accountable reporting.
What it does
Beacon AI is an anonymous reporting and case management platform with no sign-in. Users submit incidents through a guided chatbot, providing text, evidence files, and context. The system generates a Case ID and Secret Key for tracking without collecting personal data. AI-based credibility scoring evaluates reports and evidence, produces a score with explanation, and helps filter low-quality or false reports. Admins manage cases via a secure portal and communicate with users through real-time two-way anonymous messaging.
How we built it
Beacon AI is a full-stack system with a responsive frontend, chatbot, backend APIs, secure storage, and AI services. We used Gemini 3 Pro via Google Antigravity for development support and the Gemini 3 Flash API key across the product. The system is privacy-first with no user authentication.
Challenges we ran into
Key challenges included ensuring true anonymity with tracking, enabling two-way communication without accounts, handling sensitive evidence safely, working within cost constraints, and addressing AI and data security concerns.
Accomplishments that we're proud of
We built an end-to-end anonymous reporting system with Case ID tracking, real-time anonymous communication, explainable AI scoring, and full admin case lifecycle control.
What we learned
Anonymity needs transparency and follow-up. Evidence matters as much as text. AI must be explainable. Trust thresholds vary.
What's next for Beacon AI
We will focus on enterprise deployments, compliance tooling, and self-hosted options for governments, making reporting safe, anonymous, and actionable.
Built With
- alembic
- antigravity
- api
- axios
- css
- docker
- eslint
- fastapi
- framer
- gemini
- gemini3
- git
- github
- html
- httpx
- javascript
- lucide
- next.js
- ocr
- opencv
- passlib
- postcss
- pydantic
- pymupdf
- python
- python-jose
- radix
- react
- react-markdown
- sqlalchemy
- structlog
- supabase
- tailwindcss
- three.js
- tsparticles
- typescript
- vercel
Log in or sign up for Devpost to join the conversation.