Browse applications built on modern technologies. Explore PoC and MVP applications created by our community and discover innovative use cases for modern technologies.
Prototype SAAS SCM V4 for Indonesian fashion garment SMEs - from Batik design, Yard/Kodi control, Sritex invoice to digital marketing. AI trend forecasting, OEE 95%, ERP, KPI on Vultr Serverless Inference + AMD MI300X for cultural fashion agenda.
OreLedger pairs supplier invoices with independent assay certificates, recomputes payable metal deterministically, and routes only overbilling, grade-shortfall and duplicate exceptions to a human-signed audit ledger.
Reads steel suppliers' CBAM emissions documents in any language with Gemma on AMD, checks every value against the official EU tables with deterministic rules, and turns verified evidence into euros saved: €636 → €63 per tonne of Indonesian slab.
ProcureVoice Speak the request.We handle the paperwork. A voice-first procurement assistant that turns natural spoken purchase requests into policy-aware, approval-ready requests. Built with AssemblyAI Voice Agent API
Kaagaz turns any Indian banking-document question into an ordered, costed, regulator-aware checklist. It searches the live web, reads sources, cites every claim, and files each answer with its audit trail.
RegTrace brings company policies into AI-assisted coding. It reviews Claude’s plan before implementation, checks the actual code afterward, applies supported fixes, and performs a fresh review against the policies supplied by the user.
DevTwin is an AI-powered pre-merge change simulator that maps code impact, detects risks, finds test gaps, generates regression tests, and verifies changes before they reach production.
BlastRadius analyzes code changes, maps their real call sites, detects untested impact areas, scores regression risk, and generates focused pytest tests for the most dangerous paths.
Voice agent fairness audit tool. Matched-guise testing found 50% record failure rate for one voice, 100% female interruptions, 64-point emotion gap. Pre-deployment compliance tool for hiring, lending, insurance, etc. Built on AssemblyAI Voice Agent API.
An AI-powered hospital supply control tower that forecasts demand, detects true usable-stock shortages, and prioritizes packing, transfer, replenishment, and escalation actions before supply risk reaches the bedside.
PolicyRipple AI helps organizations assess internal changes before implementation by generating a readiness score, stakeholder impact map, tailored communications, actionable recommendations, and a phased rollout plan from a single change brief.
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.
The document reconciler lets you point at confidential paperwork, because nothing has to leave your hardware. Cross-document reconciliation on tokenised text, for piles that legally cannot go to a cloud IDP.
AI-powered document intelligence platform that cross-references contracts, invoices, and quotations to detect conflicts, assess risks, and deliver evidence-based decisions in under 90 seconds powered by AMD Instinct MI300X.
An autonomous B2B procurement gateway using local daemons and multi-agent orchestration to eliminate enterprise supply chain visibility gaps, track real-time inventory, and fully automate competitive vendor pricing negotiations.
Project Raze is an enterprise AI security platform executing surgical machine unlearning on LLMs. It ablates targeted weights to guarantee GDPR Article 17 compliance and erase copyrighted data without destroying model utility, powered by AMD & Fireworks.
GemmaForge is an enterprise AI DevOps pipeline that automates translating NVIDIA CUDA kernels into AMD ROCm HIP C++. It structurally scales 32-thread warps to 64-thread wavefronts, breaking vendor lock-in for compute-heavy sectors.
RoomKey ScopeGate is a Band-room authority gate for regulated agent workflows: it blocks sensitive context before approval, releases only scoped metadata after review, revokes access, and proves the run with receipts.
Five AI agents answer an enterprise RFP together in one Band room: planning, researching, drafting, compliance-vetoing and revising, then seal a submission-ready proposal with a hash-chained audit trail and a human approval gate.
ContextChain replaces lossy context dumps with Episodic Memory Objects (EMOs) — a structured handoff primitive that preserves decisions and provenance across multi-agent pipelines. Demoed on a 4-agent procurement workflow with live SSE streaming.
4 AI agents on Band automate enterprise expense approvals end-to-end: Budget Checker, Policy Checker (Llama 3.1 70B), Risk Evaluator, and Approval Notifier. Low-risk expenses are auto-approved in seconds; medium/high risk routes to a web admin panel.
A collaborative, multi-agent cybersecurity mesh that automates the triage, analysis, and containment of security incidents using a squad of specialized, autonomous AI agents.
A band agents turns a plain-English hardware build into a list of purchase orders — The agent does the fetching, Maths to the code, and decision for you. Provided with near Zero Hallucination risk.
4 AI agents collaborate through Band to monitor regulatory changes, analyze business impact, propose policy adaptations, and approve compliance actions with full audit trails — turning weeks of compliance work into minutes.
BandOps is an enterprise-grade command center that replaces siloed human panic with an orchestrated swarm of adversarial AI agents. Powered by Band, it resolves multi-million dollar aviation crises in seconds with immutable human-in-the-loop oversight.
Quadro CSI coordinates four Band-connected agents to turn regulated customer escalations into consent-aware, source-backed, audit-ready decision packets.
3 AI agents (Risk Analyst, Compliance Officer, Human Review) collaborating through Band for automated financial compliance with SAR filing detection and human review workflow.
ATower Of Agents is a Band-coordinated control tower where specialist AI agents screen HR candidates, retrieve evidence, flag risks, Automate the document curation, decision making, and prepare auditable human-review decision packets.
AI Charter is a multi-agent governance platform reviewing AI features before deployment. Five agents evaluate ethics, security, legal, compliance, and product impact, then debate, vote, and create transparent audit logs for responsible AI release.
FINCORD is a financial validation and fraud detection platform that helps businesses prevent financial manipulation, corruption, and irregular transactions. By combining automated data validation, pattern analysis, and real-time monitoring