Discover AI Applications

JARDO -

LazyCarts – AI Multi-Vendor Marketplace

LazyCarts is an AI-powered multi-vendor marketplace that helps sellers launch, manage, and grow online stores with intelligent product management, analytics, customer support, and modern shopping experiences.

SpecTrace AI

SpecTrace AI turns fragmented release evidence into traceable AI-powered investigations, generated engineering assets, and human approval decisions for safer software releases.

CodeGuard AI By Byte Anarchists

CodeGuard AI Studio is an AMD- and Fireworks AI-powered desktop vulnerability scanner. It allows developers to detect, review, and instantly patch security flaws in their source code files with a robust, human‑in‑the‑loop workflow.

MEIA-LAB (Multi-Agent Earnings Intelligence calls)

MEIA-LAB is a multi-agent AI platform that auto-verifies earnings call claims against SEC filings in real-time. Whisper ASR, Vision OCR, and Filing agents cross-examine audio, slides, and 10-Qs powered by AMD GPUs cutting analyst research time by 90%.

OmniCore — Zero-Token-First Hybrid Routing Agent

OmniCore uses a 5-layer cost ladder to minimize Fireworks tokens: rule-based math solver, 60+factual lookups,and Gemma 2B local inference handle 12 of 20 tasks at zero API cost. Only code and logic hit Fireworks. Estimated2,100 tokens,87–90% accuracy

AI Smart Warehouse Safety & Audit System

An end-to-end vision & analytics pipeline automating safety & compliance auditing in logistics. Using fine-tuned Qwen-2.5-VL on AMD Instinct hardware, it turns manual warehouse oversight into a proactive, real-time intelligence layer.

AgeBand

AgeBand — privacy-first passive age assurance for chat. An LLM infers an age band from how you write; a deterministic shell decides band, confidence, and safety posture, so it can't be confidently wrong. On-prem on AMD MI300X/ROCm/vLLM.

OFF-Guard

OFF Guard is an AI-powered KYC (Know Your Customer) compliance screening assistant

Dobbies

Dobbies is an automated security auditor for AI Agents — combining LLM-powered static analysis with AI-vs-AI attack simulations to detect OWASP LLM Top 10 vulnerabilities before they reach production. Pre-deployment, proactive, actionable.

Saboteur: Chaos Monkey for AI agents

Saboteur breaks your AI agent on purpose. It injects fake API errors, rate limits, bad tool data and lost context into a running agent, then scores how well it recovers. Chaos Monkey but for AI agents.

NeuroBridge: AI Fraud Triage & Smart Routing

An AI-powered security layer for L2-to-fiat gateways. It leverages AMD compute for real-time anomaly detection and Fireworks.ai for smart, token-efficient routing, giving merchants instant, human-readable fraud alerts.

SentinelAI – AI Cybersecurity Threat Detector

SentinelAI is an AI-powered platform that analyzes security logs, detects threats, maps them to MITRE ATT&CK, and generates incident reports through an interactive dashboard and AI chat assistant.

OpenRisk AI

OpenRisk AI is an AI-powered smart contract security copilot that analyzes Solidity repositories from GitHub, performs automated static analysis, calculates security risk scores, and generates actionable audit reports using large language models.

AuditAgent: AI Smart-Contract Oracle

Autonomous AI agent auditing Solidity smart contracts via LLM reasoning on AMD MI300X GPUs (Fireworks AI). Writes immutable risk verdicts on-chain via FETCH-REASON-SCORE-ATTEST loop.

PayEvery

PayEvery is an AI-powered payment bridge enabling 95M+ mobile wallet users in Bangladesh to generate one-time virtual dollar cards for global tech purchases, protected by real-time scam detection powered by Google Gemma 4 and AMD compute.

LabPiolet AI

LabPilot AI is an AI-powered laboratory safety assistant that identifies chemical, biological, and physical hazards, provides safety recommendations, and generates risk assessment reports to improve lab safety.

OlfactAI - AI Smell Intelligence Agent

OlfactAI detects dangerous smells like LPG gas leaks, smoke, and spoiled food using AI — providing real-time bilingual emergency instructions in Bahasa Indonesia and English. Because your nose can't always protect you. Ours can.

Token-Efficient Hybrid AI Gatekeeper

An intelligent, Dockerized hybrid AI proxy gateway that dynamically routes prompts between local Ollama edge nodes and Fireworks AI cloud tiers based on task complexity to minimize token spend.

FusionNet — Secure Federated Fine-Tuning for LLMS

FusionNet is an AMD-optimized federated AI platform enabling edge devices to collaboratively fine-tune LLMs without sharing raw data using LoRA and differential privacy. We don't move data to the model. We move the model to the data.

Aegisflow Security dashboard

AegisFlow: real-time network security for banking SOC teams. Live packet capture, AI-driven threat triage, and tamper-proof forensic evidence — built by a student, engineered like production.

Redactify - AI powered Document and Image Redactor

Redactify is an AI-assisted document and image redaction tool designed to help users permanently remove personally identifiable information (PII), such as names, emails, and addresses, from PDFs and images. The application is powered by Fireworks AI.

Marrow: Where FinOps Meets SecOps

An AI-powered command center correlating AWS Cost Explorer data with Security Hub findings. Powered by DeepSeek via Fireworks AI on AMD Instinct MI300X, Marrow optimizes cloud spend and reduces attack surfaces simultaneously.

Project Sentinel

Real-time threat intelligence combining live global events, AI analytics, and automated workflows to proactively mitigate operational risks.

CrimeSpots: Live Crime Awareness for Cape Town

A live map of Cape Town that reads public crime reports with AI and grades each suburb red, amber or blue. Check it before you head out. Free, privacy-first, and built to work on cheap phones and bad signal.

VisionLedger

VisionLedger is a revolutionary AI verification platform that analyzes the environmental evidence like tree plantations, solar installations, waste management and other sustainability initiatives in form of photos and documentation.

GemmaJudge: Gemma Attacks and Judges Itself

One open-weight Gemma family generates adversarial prompts and judges target responses, with case-level evidence, a CI regression gate, and committed self-hosted AMD MI300X proof via vLLM + ROCm.

Project Raze: Neural Decontamination Platform

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.

RDNA AI Guard

Unified, 100% local real-time AI triage and parallel multi-agent cloud escalation bridging network edge intrusion and financial ledger fraud.