Browse applications built on modern technologies. Explore PoC and MVP applications created by our community and discover innovative use cases for modern technologies.
RepoAlpha is a Bloomberg Terminal for Open Source M&A. It uses Bright Data scrape who is starring GitHub repos, detects corporate adoption signals from Nvidia, Google and Anthropic engineers, and generates BUY/SELL ratings before it becomes public news.
A voice-to-voice AI translation layer that helps customer support teams serve users in any language, dialects, and jargon combining real-time translation, company agents, local knowledge bases and human-in-the-loop quality control.
small python web service and vue app to enable realtime translation/transcription of streamed audio
WisdomAI revives ancient wisdom securely with AI by making centuries-old knowledge accessible to today’s generation in any language. Users can upload wisdom texts, ask questions, choose the answer language, and receive responses grounded from k-base..
DataGuard is a secure enterprise RAG platform with built-in prompt injection detection, audit logging, and real-time token inspection powered by Lobster Trap so every document query is grounded, safe, and compliance-ready.
Meridian turns any video into a queryable knowledge base. Upload a video, ask a question in plain English, and get a precise timestamped answer drawn from speech, on screen text, and what was visually happening at that exact moment.
PHRAXIS turns a spoken developer idea into a production-ready GitHub PR in under 4 minutes. IBM Bob reads the full codebase. IBM Quantum QAOA finds the optimal change path. IBM Watson STT and NLU extract structured intent. Voice to impact, end to end.
BobBridge transforms natural language prompts into live mock API endpoints and framework-specific code scaffolds using IBM watsonx.ai. Designed for rapid prototyping, it generates contracts and provides a seamless handoff to IBM Bob for implementation.
VoiceLens is an AI-powered speech confidence coach that records your voice, transcribes it live, and delivers a detailed analysis of fluency, filler words, pace, clarity, and whether you should seek therapy.
CodeAtlas is an AI-native developer intelligence platform that transforms GitHub repositories into semantic knowledge graphs with graph-aware RAG, repository intelligence, blast radius analysis, AI onboarding, and conversational code understanding.
Dana Whisper is an AI-powered real-time meeting assistant that transcribes conversations and provides contextual guidance during meetings. IBM BOB helped us design a scalable multi-agent architecture powered by Watson and watsonx.ai.
BilimForAll is a gamified "Duolingo for professions" powered by a multi-agent Gemini 1.5 system. It features automated onboarding, content generation, voice roleplays with AI mentors, and Context Caching, slashing enterprise API costs by up to 85%
An autonomous, multilingual security copilot powered by Gemini 2.5 and Veea Lobster Trap. It acts as an interactive incident response agent, executing live server mitigation patches and sending reports to engineers over real-time telecom voice calls.
VaultIQ is a multilingual Visual RAG platform that lets enterprises query image-based documents in Arabic or English — with grounded, page-cited answers and a built-in AI governance layer for full auditability.
MindShield by BrainSNN is a 3D cognitive firewall for enterprise AI, using Gemini analysis, Veea Lobster Trap, and self-evolving defenses to detect manipulation, block unsafe prompts, and fuse multimodal knowledge.
Ken hears every word. You understand every one. Ken is a free, self-hostable, open-source AI co-listener that gives real-time, explainable, personalized support during immigration, legal, and medical conversations — built on AMD ROCm.
LeadOS is a white label agentic AI platform that engages leads through chat or voice, qualifies them with business context, scores intent, and routes the best opportunities to human teams.
Enhanced Hindi support in Qwen3-Omni-30B-A3B-Instruct using AI4Bharat datasets (ShrutiLipi, Rasa, Lahaja,QnA) and a custom LLaMA Factory variant for Hindi audio understanding. Audio output training was limited due to lack of official codec support.
YuhoLens is a 14B open-weight Japanese-finance LLM that ingests EDINET yuho filings and drafts cited equity memos through a 4-agent LangGraph composer, shipped with a reproducibility ledger and tuned for AMD ROCm inference.
self-evolving multilingual cyber immune system that detects threats in 40+ languages, synthesises defences through adversarial AI battle, formally verifies with Z3, broadcasts immunity across an encrypted P2P mesh-every connected org inherits protection.
Forma is a Chrome extension that catches vague UI vocabulary in vibe coders' prompts on v0, Cursor, Lovable, and Bolt — rewriting them with canonical component terms. 7 agents on AMD MI300X: Llama 3.1 8B for real-time scoring, 70B AWQ for Memory Engine.
The AMD Multimodal Workbench is a unified app featuring three specialized modes: industrial defect inspection, an educational medical imaging workflow, and a vision-language assistant. It leverages cutting-edge multimodal models on AMD ROCm. 4
HIP-Hop is a multi-agent AI swarm that autonomously translates legacy NVIDIA CUDA code into AMD ROCm/HIP C++ syntax, eliminating vendor lock-in using local MI300X compute.
Upload complex documents — PDFs with charts, tables, scans — and have a natural languageconversation about everything, including the visuals.
Voice-first AI triage agent that catches atypical cardiovascular presentations standard ER triage misses, especially in underserved populations.
SketchTeach is an AI-powered tutor that transforms complex concepts into synchronized Manim whiteboard animations with voice narration. Powered by Qwen2.5 on AMD MI300X, it automates the lesson-creation process through a unique multi-agent pipeline.
LinguaLink is a real-time AI voice & text translation desktop app — speak in one language, hear the reply in another, powered by Deepgram ASR, AMD Fireworks AI serverless model API, and Deepgram TTS.
Miwa is a real-time Discord voice overlay that transcribes Japanese speech, shows instant English translation, refines tone with a 70B LLM, generates contextual AI replies, and supports bot voice/chat delivery with sub-800ms target latency on AMD MI300X.