Bobby rajak@Bobby05

Aria AI Voice Assistant is an speech, processes their requests using AI, and responds with real-time voice answers. It can help with questions, conversations, and everyday tasks through a simple hands-free interface. 🎙️🤖

PocketForge AI is an autonomous AI Dark Factory testbed and simulation pipeline designed to build, stress-test, self-repair, and cryptographically verify mission-critical software without human intervention. Governed by the core philosophy "AI that builds software, breaks it, fixes it, and proves it," the system automates the entire software lifecycle inside a deterministic closed loop. The architecture operates through a 6-agent orchestration pipeline using strongly typed handoffs: Planner Agent: Decomposes ingress task briefs into acceptance criteria, failure hypotheses, and edge-case models. Architect Agent: Defines database schemas, state machines, and mathematical conservation invariants. Implementation Agent: Synthesizes backend business logic and integer minor-unit financial accounting. Adversarial Test Agent (Red Team): Executes a 50-vector automated exploit suite targeting race conditions, duplicate retries, and network cutoffs. Repair Agent: Diagnoses failure traces and injects atomic mutex locks and rollback invariants. Verification Agent (Blue Team): Acts as an independent zero-trust auditor, validating ledger consistency and generating sealed SHA-256 evidence proof hashes. Its reference demo engineers an ultra-resilient wallet transfer engine guaranteeing zero fund loss, race-condition immunity, and instant idempotency across thousands of concurrent attacks.

Decision Memory AI is an AI-powered developer knowledge system designed to solve a common problem in software development: developers can understand what the code does, but often struggle to understand why a technical decision was made. Important context is usually scattered across Git history, Pull Requests, tickets, issues, and documentation. Finding the reasoning behind an architectural or technical decision can require manually searching through multiple sources, asking senior developers, and revisiting old discussions. Decision Memory AI brings this information together into an evidence-backed Decision Graph. A developer can ask a question such as: “Why did we choose OAuth instead of JWT?” The system analyzes relevant project information using specialized AI agents for sources such as Git history, Pull Requests, tickets, and documentation. It then connects the extracted evidence into a structured Decision Graph. Instead of returning a generic AI response, Decision Memory AI presents: The decision that was made Why the decision was made Alternatives that were considered Rejected options Relevant context Decision timeline Supporting evidence and citations A key feature is our evidence-first approach. The system is designed to avoid inventing unsupported PR numbers, commits, dates, people, or technical decisions. When sufficient evidence is not available, it can indicate that the available information is insufficient rather than presenting an unsupported answer as fact. The Decision Graph provides a visual and structured way to understand how a technical decision connects to its supporting evidence and historical context. This can help with developer onboarding, application maintenance, architectural understanding, and preserving valuable project knowledge. IBM Bob 2.0 IBM Bob 2.0 was used as a core part of the development workflow. IBM Bob IDE was used for repository analysis, architecture understanding, implementation planning, development, debugging, testing, and code review. The runtime application also uses specialized AI agents to analyze project sources. These application agents are separate from IBM Bob's role as the development partner used to build and improve the project. Technology React TypeScript Tailwind CSS Node.js Google AI Studio / Gemini Decision Graph visualization Git, PR, ticket, and documentation-based evidence Decision Memory AI turns scattered development history into searchable technical memory — helping developers understand not only what the code does, but why it was built that way.

FinGuard AI is an AI-powered financial risk intelligence platform that helps businesses detect fraud, forecast cash flow, analyze business risks, and make data-driven financial decisions. The platform provides five core capabilities: Fraud & Anomaly Detection, 30/60/90-Day Cash-Flow Forecasting, Business Risk Intelligence, AI CFO Assistant, and AMD GPU/ROCm acceleration. The Fraud Detection module identifies suspicious transactions and provides explainable risk factors. Cash-flow forecasting analyzes financial trends to estimate future revenue, expenses, and balance. Business Risk Intelligence evaluates customers, suppliers, payments, expenses, and revenue to identify potential risks. The AI CFO allows users to ask natural-language questions such as “Why is my cash flow decreasing?” or “Which transactions are suspicious?” and receive insights based on the available financial data. For computationally intensive AI/ML workloads, FinGuard AI is designed to use AMD GPUs with ROCm where supported. The platform also provides GPU monitoring for utilization, VRAM, temperature, power, workload, and ROCm status. Technology: Next.js, TypeScript, Tailwind CSS, Python, FastAPI, PostgreSQL, PyTorch, Scikit-learn, Pandas, NumPy, AMD GPU, and ROCm. FinGuard AI transforms complex financial data into clear, actionable intelligence—helping businesses understand what is happening, why it is happening, and what could happen next.

ARIA is a real-time, browser-based voice assistant built entirely on AssemblyAI's Voice Agent API — a single WebSocket connection that handles speech-to-text, LLM reasoning, and text-to-speech together, so there's no separate pipeline to stitch together by hand. Click "start talking," and ARIA listens through your microphone, transcribes your speech in real time, thinks through a response, and speaks back — all while a living, organic orb on screen reacts to the actual audio. It breathes gently at rest, brightens and pulses with your voice while you talk, shifts color while thinking, and moves with the AI's own voice while it replies. A scrolling live transcript runs underneath, so the full conversation stays visible and re-readable. Under the hood, a lightweight Next.js server route mints a short-lived, single-use AssemblyAI token per session, so the real API key never reaches the browser. A custom AudioWorklet captures raw microphone audio directly on the audio thread, buffers it into fifty-millisecond chunks, and converts it to sixteen-bit PCM at twenty-four kilohertz — the exact format the Voice Agent API expects — with zero external audio libraries. A single WebSocket streams that audio up and receives transcript deltas, agent replies, and streamed reply audio back down, handling turn-taking and interruptions natively. Reply audio is decoded and queued on an AudioContext timeline for gapless playback. The interface is hand-built: drifting, blurred SVG circles merged with an SVG goo filter form the "living" orb, driven by real amplitude data read straight from the mic and playback streams, so it always reflects what's actually happening acoustically rather than a canned animation. Most voice-AI demos bolt together three separate APIs with noticeable lag, or hide behind a static spinner with no real feedback. ARIA leans on AssemblyAI's managed pipeline for low-latency, natural conversation, paired with an interface that feels genuinely alive.

Six Matches. One leaderboard. Build on AMD, earn points and compete for prizes.

🌎 Hybrid Hackathon | 💻 Online Build: 3–8 November 2026 | On-site Phase: 8 November 2026 📍 On-site Venue: Pierpont Place 163 W Pierpont Ave, Salt Lake City, UT 84101, USA ✨ Please note that travel and accommodation are not covered.

🌎 Hybrid Hackathon | 💻 Online Build: October 16–19, 2026 | 📍 On-site: October 19–20, 2026 The official hackathon of AI & Big Data Expo Europe, organized by TechEx Events and lablab.ai. Build AI applications and agents across AI, cybersecurity, cloud, IoT, automation, edge computing, digital transformation, and data infrastructure, then showcase your project live at one of Europe's largest enterprise technology events. 📍 On-site Venue (October 19–20): RAI Amsterdam Europaplein 24, 1078 GZ Amsterdam, Netherlands 📍 If approved for the on-site experience, you'll also have the opportunity to showcase your project live. Travel and accommodation will not be covered.

🌎 Hybrid Hackathon | 💻 Online Build: 12-17 October 2026 | On-site phase: 17-18 October 2026 To participate in the hackathon, simply click Sign up with AMD. If you are not already a member of the AMD AI Developer Program (ADP), you will need to create an account. The AMD Developer Hackathon: ACT III is a hands-on event for developers, founders, engineers and builders who want to push what’s possible with AI on real infrastructure. 📍 If approved for the on-site experience, you'll also have the opportunity to showcase your project live. Travel and accommodation will not be covered.

🌎 Online Hackathon | 💻 Online Build: Sept 26 - Oct 5, 2026 A virtual, global hackathon. Build from anywhere in the world. Dark Factory is the official hackathon of WeAreDevelopers World Congress North America, presented by BAND, but you don't need to be in San Jose to take part. For one week, you build a software factory in BAND Desktop: a band of coding agents that plans work, implements it and checks its own results. 🎟️ Bonus for sign-ups: you are eligible for a free ticket to WeAreDevelopers World Congress North America in San Jose (Sep 23–25). If you do go, stop by the BAND booth: a surprise is waiting for you.