Browse applications built on modern technologies. Explore PoC and MVP applications created by our community and discover innovative use cases for modern technologies.
Aura is a sovereign, neuro-symbolic AI operating system that turns natural-language intent into transparent, governed action through local parsing, agent collaboration, bounded experimentation, verifiable learning, and human-controlled decision gates.
NarraFlow is a multi-agent AI storytelling platform that transforms simple prompts into immersive stories. Specialized AI agents collaborate to plan plots, develop characters, maintain consistency, and generate engaging narratives in real time
An agentic pipeline that takes any CUDA repo and autonomously ports it to ROCm/HIP — analyze → transpile → compile → test → benchmark on real AMD GPUs — with an LLM repair loop fixing everything the mechanical tools can't.
Taskly classifies incoming tasks into 8 categories (QA, math, code, NLP) and routes each to the optimal Fireworks AI model with a tuned prompt — maximizing accuracy while minimizing token usage in a fully Dockerized pipeline.
3DP Agent analyzes STL files before printing, catching wall-thickness failures and unsupported overhangs invisible to the eye. Five specialized AI agents vote on a consensus printability score, with inference running on AMD ROCm via vLLM.
Vision-based video captioning pipeline using Gemma 4 31B. Analyzes video frames,nbuilds scene understanding, generates 4 caption styles with automatic quality scoring.
A context-aware Windows window manager: say what you're doing in plain English, and an AMD-hosted LLM instantly rearranges your open windows around your intent.
Vektor is a deterministic Guardian Runtime that intercepts autonomous AI tool calls before execution. Because it evaluates actions against enterprise policies without language models, it prevents unsafe behaviour and logs verifiable proof.
ReproForge Sentinel evaluates AI/ML claims against declared evidence and policies, generates transparent risk and reproducibility scores, and seals the findings into an auditable Reproducibility Passport with explicit proof gaps.
SWARM (Swarm Workforce Autonomous Refactoring Migration) Autonomous CUDA-to-ROCm porting pipeline with a self-healing repair loop — verified on real AMD hardware, zero fabricated numbers.
Zero-trust enterprise privacy middleware for LLMs. AegisLayer uses AMD ROCm-accelerated NER to redact sensitive PII locally before cloud inference, and seamlessly restores original values in the response using an ephemeral in-memory vault.
Release any message, a rumor, an idea, a product claim, into a small world of AI minds living on one AMD MI300X. Murmur measures what society does to it: spread, mutation, belief, doubt, live web research by the minds, and time to reach authority.
Type a policy. Watch a thousand minds panic, steal, marry, riot, in real time, then explain themselves in plain English. Three Gemma 4 models, five psychological theories, one AMD MI300X: a civilization that governs itself the instant you govern it.
init.dev analyzes your GitHub profile and any open source repository to match you with the right issues. It generates personalized investigation guides, highlights the most relevant files, and helps developers make their first contribution faster.
Most AI advisors hand you a confident paragraph and call it a day. Chronos builds a memory graph from your GitHub, docs, and real evidence, then runs a Multi-Agent Council to simulate your next big decision with probability scores and Expected Regret.
SchemaIntern converts natural language questions into accurate SQL queries by understanding your database schema. Connect any PostgreSQL database, ask questions in plain English, and get safe, optimized SQL instantly (no SQL knowledge required.)
An autonomous, multi-agent platform that scans, migrates, validates, and optimizes CUDA applications to AMD ROCm. Features tree-sitter AST parsing, Gemma-based LangGraph translation, sandboxed Docker compilation, and a real-time Next.js dashboard.
Transform your codebase into a VC-grade startup pitch deck in seconds. Commit-To-Pitch analyzes your code, extracts technical intent, and generates a compelling, investable pitch using advanced AI.
Lifeline AI eliminates the "Context Tax" by providing a platform for seamless conversation portability between models. Users can move chat histories across ChatGPT, Claude, Gemini, and DeepSeek with one click and zero context loss.
CodeSentinal is an AI-powered code review and security analysis platform that automatically detects bugs,vulnerabilities, code smells, and performance issues while providing intelligent suggestions to help developers write secure,high-quality code faster.
Point it at any CUDA-first PyTorch repo: it scans blockers, plans with a multi-model agent orchestra, auto-patches, builds a ROCm container, validates on real AMD hardware, and returns an honest before→after AMD-readiness score.
AI-powered security scanner that detects code vulnerabilities, generates production-ready fixes, and scans Docker containers for CVEs — all in one tool.
Smart-Router Agent answers all 8 Track 1 task categories at a fraction of the token cost. A free on-device model handles the easy tasks; the paid Fireworks API is reserved for hard math, logic, and code.
We built ReCodeX, the Legacy Code Modernization Engine. It's an AI platform that automatically transforms legacy "spaghetti" code into clean, secure, and cloud-ready infrastructure. Powered by AMD high-performance cloud hardware.
A typed, decaying, graph-aware memory layer for software engineers that runs fully local on AMD ROCm / Instinct GPUs — no cloud, no data leaves the machine.
An AI-powered assistant that analyzes GitHub repositories and generates project summaries, marketing copy, and growth strategies for founders and developers
Local-first Rust proxy that intercepts AI agent tool calls over MCP, blocking destructive commands and obfuscated attacks via static policy + semantic LLM evaluation on AMD ROCm, with SQLite audit logging. Zero data exfiltration. 34 tests, containerized.
Clinical decision support for elderly polypharmacy. A deterministic engine 102 drug interactions,61 AGS Beers rules, 58 STOPP/START v3 criteria runs in ~3ms on every request, gated on age and eGFR, escalating to a cloud LLM only when the case warrants it.
Cognexa is a hybrid AI routing platform that intelligently classifies user requests, executes simple tasks locally, and escalates complex queries to Fireworks AI only when needed—reducing latency, API costs, and unnecessary token usage.
Dominate high-stakes negotiations with Kifani AI. Leveraging Fireworks AI and the Gemma model, we provide real-time "Power Move" scripts and tactical blind-spot analysis for salary talks, VC rounds and legal closings. Negotiate like a pioneer.