Browse applications built on modern technologies. Explore PoC and MVP applications created by our community and discover innovative use cases for modern technologies.
RoboProof trains and evaluates a VEX robot’s motion policy in the original Simulator/iraLIB, combining GVF path following with TD3 residual learning on AMD GPUs.
InnerOS turns AMD-powered AI into governed real-world action. Gemma reasons, Evolus runs workflows, and Alexa/Inner Oz can query live operations and create governed GitHub tasks for AI agents, with approvals, evidence, and replay.
FactoryProof is a BAND-native software factory that plans, builds, verifies, and gates changes with observable evidence. The reference Pocketful track demonstrates a money-safe wallet transfer.
TrustForge is an autonomous software factory that builds, challenges, verifies, repairs and proves critical software behavior before it ships.
A BAND team using Codex and Claude built and independently reviewed a four-stage wallet and payments app. The Docker services pass 193 checks, with reusable seat mandates and an authentic room export documenting a supervised practice run.
Tablekeeper is a smart restaurant reservation platform that simplifies restaurant discovery, real-time table availability, and booking management through an intuitive interface and reliable reservation workflows.
Seatright is a software factory on Band. Five AI agents, set up like a rock band, built all four Tablekeeper stages from one message. They planned, coded, reviewed and shipped it among themselves. The video is best viewed with sound on!
A band of four AI coding agents that builds a Venmo-style wallet and attacks its own work for double-spends and race conditions before accepting anything.
Four agents on BAND (Claude, MiMo, local Qwen 80B) built Pocketful, a wallet that never creates, destroys or double-spends money. Custom band-sdk adapter, double-entry ledger, scoped idempotency and 82 passing tests, coordinated in one shared room.
Ripcord is a reliability layer for voice agents that detects uncertain or inconsistent critical information, pauses risky workflows, and asks targeted clarification questions before a misunderstanding becomes a real-world action.
Plant operators walk their inspection rounds and just talk. RONDA logs every reading, checks it against the equipment's real limits on the server, and raises the work notification before the shift ends.
NovaFuse is a voice-first AI study partner & STAR mock interview studio. Practice active-recall quizzes and technical interviews through real-time bidirectional voice conversations with adaptive difficulty and automated weak-topic tracking.
Codebox AI is a voice-first coding assistant for blind and disabled developers. Just speak to explore your codebase, ask questions, and make code changes, no screen or mouse needed.
It is an AI which finds code languages and then like finds mistake
Changewise turns dependency upgrades into proven migrations: it reads release notes, finds affected code, IBM Bob fixes it, and test-backed receipts prove it. Demo: 33/33 → 3/33 after version bump → 33/33, zero test files changed.
Architecture Storyteller 2.0 and Ripple-Agent is a VS Code and IBM Bob IDE extension that maps polyglot codebases, visualizes dependencies, predicts change impact, generates documentation, and creates architecture dossiers.
AI copilot that turns any GitHub repo into a personalised onboarding journey: auto repo analysis, day-by-day plans, codebase chat, code validation, first-task suggestions and PR drafts. Built with IBM Bob, powered by IBM watsonx Granite.
RepoPilot is an agentic AI onboarding platform powered by IBM Bob 2.0. By simply pasting a GitHub URL, a new developer receives an autonomous architecture analysis, an interactive 8-step code tour with line-level citations, and a grounded Q&A interface.
An adaptive, evidence-first code review system that investigates software changes, selects testing strategies based on evidence gaps, produces auditable receipts, and converts verified bugs into regression protection.
AI-powered API migration guard and root-cause analysis agent built on IBM Bob 2.0 that eliminates cascading CI failure noise and verifies non-breaking migrations.
Sethu brings API migration into IBM Bob. It traces breaking OpenAPI changes into consumer code, helps repair affected integrations, and checks recorded evidence so developers can assess whether each change has been addressed.
Regression Whisperer turns messy stack traces into a searchable, AI-distilled fix library for your whole team, paste an error, get instant matches from past fixes. Includes voice input, CSV bulk-upload, and a full multi-tenant analytics dashboard.
automates diagnosing and fixing API contract inconsistencies. It orchestrates ten workflow stages, from bug intake to verified resolution, recording every step in a persistent Contract Evidence Graph for a complete, auditable chain of evidence.
CodePulse automates code review with parallel AI analysis agents, documentation-aware checks, regression-test generation, and validation in one developer workflow.
Bob Guardian is a multi agent auditing system that detects when an AI coding agent's project context and documentation drift from the actual code, surfacing contradictions with supporting evidence before they cause downstream bugs.
A CLI-first infrastructure intelligence layer that connects software and AI workloads to physical telemetry, attribution, resource accounting, and evidence-grounded AI insights inside the developer workflow.
PhysicsReview is a physics-aware code reviewer for simulation software. It maps code to physical laws, generates physics tests and flags the bugs that don't crash but give wrong answers: unit mix-ups, missing constants and unstable time steps.
Yazi Dev is an AI-powered developer assistant that helps students and developers understand unfamiliar codebases, debug issues, identify testing gaps, and prepare for deployment using their actual project context.
CodeGuard AI identifies code-risk hotspots, investigates bugs, independently verifies fixes with real tests, and builds reusable memory from only verified debugging incidents.
CodeGuard AI is an AI-powered code security assistant that detects bugs, vulnerabilities, and insecure coding patterns, explains the risks, and provides actionable fixes to help developers build safer software.