Shreekumar Shah@shree-shah

VoteWise is a comprehensive, multi-modal civic education platform that combines AI-powered conversational learning with interactive simulations, gamified quizzes, and geospatial tools to make India's election process accessible to every citizen.

PriceGhost is a full-stack forensic intelligence platform that detects, measures, and cryptographically proves dynamic geographic pricing discrimination. THE PROBLEM: Corporations silently charge different prices based on your location, device, and browser fingerprint. 78% of consumers report feeling targeted by location-based pricing bias, yet proving it is nearly impossible. HOW IT WORKS: PriceGhost coordinates 10 simultaneous residential proxy scrapes across global coordinates (Mumbai, New York, London, Tokyo, Berlin, Sydney, Lagos, Buenos Aires, Dubai, Singapore) via Bright Data's Web Unlocker API. Each scrape rotates device fingerprints and captures raw HTML payloads. STATISTICAL FORENSICS ENGINE: Four custom mathematical algorithms run natively — Gini Coefficient of Spatial Inequality, Coefficient of Variation, Mann-Whitney U Significance Test (p < 0.05), and GDP Pearson Wealth Correlation — establishing courtroom-ready mathematical proof of pricing discrimination. AI-POWERED PARSING: When standard regex price extraction fails on complex HTML, Featherless AI's hosted Llama-3 model acts as a semantic fallback parser. AI/ML API generates authoritative natural language indictments styled as investigative exposés. COGNITIVE MEMORY: Cognee's semantic graph database indexes every pricing anomaly, enabling live queries against historical precedents to expose long-term corporate discrimination patterns. AUTOMATED ALERTS: TriggerWare webhooks automatically dispatch incident alerts to legal networks when Gini/Pearson indices flag "Severe" exploitation levels. EVIDENCE INTEGRITY: Every scrape result is sealed with SHA-256 cryptographic signatures and timestamp chains, producing immutable evidence packages exportable as courtroom-ready JSON dossiers. BUILT WITH: Next.js 16 (Turbopack), better-sqlite3 (7-table schema with WAL), Recharts composed visualizations, Leaflet dynamic trace maps.

The Agentic Black Box: $4.8M is the average cost of a production AI failure because enterprises deploy non-deterministic agents blind. Traditional software fails predictably, but AI agents fail probabilistically. When an agent hallucinates a variable or fails to call a tool, it confidently carries that corrupted data downstream, infecting the entire pipeline. By the time anyone notices, the damage is done. The Solution: Deterministic Chaos Inspired by Netflix’s Chaos Monkey, HEXFIRE is the world's first Chaos Engineering platform purpose-built for autonomous AI agents. We deliberately break your agent workflows to show you exactly where they crumble. Instead of guessing, HEXFIRE lets you execute targeted strikes. You can inject 6 deterministic fault scenarios into any step of your pipeline: Hallucinations, Tool Timeouts, Adversarial Prompts, Context Corruption, Latency Spikes, and Permission Revocations. Technical Gravity (Beyond the Wrapper) Remove ALL LLM calls from HEXFIRE and you still have four custom, independent computer science systems. The AI enhances HEXFIRE; it doesn't define it: DAG Cascade Analysis: Agent pipelines are Directed Acyclic Graphs (DAGs). When a fault hits one node, HEXFIRE mathematically traverses the downstream edges to determine the "blast radius." Deterministic Resilience Scoring: We compute a hard, mathematical ★-to-★★★★★ safety rating based on Survival Rate and Fault Isolation. Cryptographic Proof: Every test execution is hashed into a SHA-256 tamper-evident compliance audit chain. Generative Forensics: Gemini 2.5 Flash analyzes the wreckage and generates an exportable forensic report explaining exactly what died and how to patch it. Sponsor Integration HEXFIRE utilizes Vultr Serverless Inference for ultra-low latency execution and massive concurrency during chaos testing spikes. We leverage Google Gemini 2.5 Flash for deep resilience analysis, forensic reporting, and robust 503 capacity fallback chains. Built solo by Shree Shah.

Every year, enterprises burn $2.1 trillion in stalled initiatives — not because the strategy was wrong, but because the presenter couldn't survive 12 minutes of hostile cross-examination. Board proposals face a 74% first-hearing rejection rate. No tool exists for adversarial rehearsal. Until Foxhole. WHAT IT DOES: Foxhole is an AI boardroom war room where executives rehearse proposals against adversarial AI directors — each with distinct biases and cross-examination instincts — powered by Gemini 2.5 Flash. SETUP: Define your proposal (M&A deal, budget, tech initiative). Configure a margin-obsessed CFO, skeptical CTO, and compliance-hardened Lead Director. Choose rigor: Friendly, Standard, or Stress-Test. WAR ROOM: Board members challenge you in real-time with streaming AI dialogue. A shadow engine tracks Board Approval Rating (0–100%), per-member alignment, Risk Vulnerability Feed (Financial/Technical/Compliance), and tactical coaching tips. DEBRIEF: Radar chart analysis, auto-generated SWOT matrix, and print-ready PDF risk report. SECURITY — NOT JUST A CHATBOT: Board proposals contain undisclosed M&A targets and financial projections. Foxhole integrates Veea's Lobster Trap DPI proxy to audit ALL outbound prompts for PII leakage and prompt injection — before tokens reach the AI. Graceful fallback to direct encrypted API when proxy is offline. ARCHITECTURE: • Next.js 16 + Turbopack • Gemini 2.5 Flash: streamText() for dialogue + generateText() for structured JSON assessment • Recharts polar radar visualization • Glassmorphic dark theme + Framer Motion animations • Veea Lobster Trap DPI with graceful degradation ChatGPT agrees with everything. Foxhole fights back. Every weak argument exposed, every financial gap exploited — BEFORE the real boardroom.

The Problem: AI coding agents are bleeding engineering budgets dry. Industry data shows that 43% of AI-generated code requires manual rewrites, wasting 11.4 developer hours per week. Why? Because AI agents operate in a vacuum. Every interaction wastes tokens re-discovering your architecture, naming conventions, and unwritten tribal knowledge. The Solution: AgentIQ is an enterprise-grade context synchronization layer that bridges the gap between human engineering intent and AI execution. Powered by IBM Granite 3.1, our platform executes three core pillars of AI governance: 1. AUDIT: AgentIQ recursively scans your GitHub repository, using IBM Granite to analyze five dimensions: Naming Conventions, Architecture, Code Patterns, Build/Deploy pipelines, and Documentation. It extracts unwritten "Tribal Knowledge" from PR reviews and calculates your overall "AgentIQ Score." 2. HEAL: The platform instantly generates optimized context files for every major AI coding agent. For this hackathon, we built deep, native integration for IBM Bob. AgentIQ generates customized .bob/modes/agentiq-optimized.yaml (Custom Modes) and .bob/skills/context-sync.md (Custom Skills), teaching Bob your exact layer boundaries and styling rules before it writes a single line of code. 3. PROVE: AgentIQ provides quantified business value. The dashboard calculates exact ROI metrics, demonstrating how the generated context files reduce code rework by 72%, resulting in over $89,000 in annual savings for a 10-person engineering team. By transforming erratic LLMs into disciplined, context-aware development partners, AgentIQ ensures that your enterprise doesn't just use AI—it governs it.

📅 June 12–19, 2026 • June 12: Kick-off Stream • June 12–19: Online Build Phase • June 19: Project Submissions Ends 🌍 Fully online hackathon. Join and build from anywhere in the world. 🤝 Build practical enterprise workflows where AI agents communicate, coordinate, exchange context, and complete work together. 🧠 Powered by Band and Codeband, built for collaborative AI systems and agent orchestration. 🏆 Prize Pool: $10,000+ 🧑‍💻 Apply now to build the next generation of enterprise multi-agent systems.

🌍 Hybrid Hackathon | 💻 Online Build: May 25–30,2026 | 📍 SF Onsite Finale: May 30–31,2026 We’ve unlocked the web for your AI. Now build what wasn’t possible before. Build AI agents with live web data, then join the SF onsite finale at Bright Data’s new Web Data Loft. 🚀 The onsite experience will take place during the opening of Bright Data’s new SF space, The Web Data Loft. 📍 On-site Venue (May 30–31): The Web Data Loft 625 2nd St, San Francisco, CA 🏆 $18,300 cash prize + Bright Data AI Startup Program opportunities 🍕 Free food, drinks, networking, workshops, mentorship, and live demos 🎁 Onsite attendees are automatically entered into a live giveaway for 4 × $100 Amazon gift cards. 🧑‍💻 Apply now to build AI agents powered by real-time web access.

Build solutions that improve how software is built. Work with an AI that understands your codebase, helping you reduce repetitive work and build with real context. ⏳ Build your project in 48 hours. 🚀 Get hands-on with IBM Bob and explore how AI fits into real development workflows before it becomes standard. 💡 Build tools and workflows that developers would actually use. ⭐ Learn directly from experts throughout the event. 🤝 Join a team or build solo. 💰 Prize pool: $10,000 🧑🏻‍💻 Register before the Kick-Off Stream and start building with Bob.

⏱️ Build the next generation of Autonomous Agents in the heart of Europe’s AI Revolution. 📅 May 13 – 20, 2026 May 13 – 19: Online Build Phase. May 19: An exclusive on-site build day at Milan AI Week for selected participants. May 20: Awards Ceremony. 📍On-site Venue (May 19–20): Fiera Milano (Rho), Milan, Italy 🤝 Join solo or with a team. 🎟 FREE Milan AI Week conference ticket included for participants. 💰 $32,000+ prize pool 🧑‍💻 Apply now to turn your autonomous agent into a market-ready enterprise solution.