Srijan Mehrotra@srijan

SupplyPulse is an AI-powered delivery exposure platform for enterprise supply chain teams managing critical equipment deliveries across global suppliers, countries, and routes. The platform helps teams identify high-pressure deliveries, understand why they need attention, verify the external evidence behind the findings, and generate business-ready reports. For the demo, SupplyPulse is connected to the NorthBridge Energy Infrastructure workspace for the Singapore Data Center Phase 2 program. It analyzes equipment delivery records, planned need dates, forecast arrivals, source countries, destination countries, and external exposure queries. The user can ask one business question, and SupplyPulse runs an agent pipeline that checks schedule pressure, geopolitical issues, trade and tariff exposure, and route logistics concerns. Bright Data API is used to gather fresh external web evidence, a Source Quality Filter keeps useful sources, and Gemini reasons over the filtered evidence to generate summaries, recommendations, limitations, reports, and an audit trail. SupplyPulse turns static delivery tracking into an evidence-backed workflow for faster, more confident supply chain decisions.

RegulaForge is an end-to-end AI compliance enforcement platform built for regulated industries — finance, healthcare, and legal — where LLM agents must operate within strict policy boundaries. The platform works in four stages. First, compliance teams upload PDF documents such as PCI-DSS, FFIEC, or OFAC guidelines. Gemini processes these documents using structured output extraction to identify enforceable rules — conditions, actions, thresholds, and regulatory references — and compiles them into a machine-readable YAML policy loaded into Lobster Trap, a high-performance LLM proxy. Second, teams test their AI agent in a simulated finance chat environment. Every prompt passes through a two-layer enforcement pipeline: Lobster Trap handles fast pattern-based blocking, while Gemini's semantic guard catches nuanced social engineering and fraud attempts that rules alone cannot detect. Third, the red-team engine fires categorised adversarial attacks — social engineering, jailbreak attempts, tool exploitation, PII exfiltration, and sanctions fraud — against the live policy. Results stream progressively to the UI showing per-attack verdicts, enforcement layer, risk scores, and compliance explanations. Finally, when attacks expose policy gaps, the Heal module sends failing prompts to Gemini, which generates targeted YAML patches. Patches are diffed against the original policy and hot-reloaded into Lobster Trap without downtime, closing vulnerabilities in seconds. RegulaForge removes the months-long gap between writing a compliance document and enforcing it in a deployed AI system.

AgentFix is a full-stack AI agent prompt security platform built with IBM Bob IDE and Vertex AI Gemini 2.5 Flash. It solves a critical gap in AI development — developers have no way to test their agent prompts for security vulnerabilities before production deployment. The platform provides a complete 5-step pipeline: ANALYZE — Scans all 8 sections of your prompt pack across 40+ issue types. Generates a reliability score across 13 security dimensions with severity ratings and actionable fix recommendations. ATTACK MODE — A two-lane red team engine. Lane 1 uses Gemini 2.5 to generate intelligent context-aware attack scenarios mapped to 3 simulation profiles and 14 named test IDs. Lane 2 pulls from 600+ real adversarial payloads from the HuggingFace prompt injection dataset, scored by complexity and selected with hard-payload bias. Supports multi-turn attacks up to 3 conversation turns. Live Agent Probe mode fires attacks at a real Vertex AI agent and captures actual responses as evidence. SELF-IMPROVE — Iterative prompt hardening engine that reads every failed attack, understands the root cause, and automatically rewrites the vulnerable sections. Runs up to 5 improvement iterations until a reliability threshold is met. VERIFY IMPROVEMENT — Reruns the exact same attacks against the improved prompt. Shows before and after reliability scores, failure count reduction, and vulnerability reduction percentage. REPORT — One-click developer report exported as PDF covering all issues, attack results, improvements made, and remaining risks. Built entirely with IBM Bob IDE as the primary development partner. Full session documentation available in the repository. Powered by Vertex AI Gemini 2.5 Flash via streamGenerateContent SSE with built-in observability, retry logic, and Express Mode support.

Welcome to Dr.Chat - the AI-powered medical chatbot app that connects patients with doctors in a seamless and intuitive way. Our backend is built using FastAPI and leverages the state-of-the-art AI21 LLM model, trained on a massive dataset of actual conversations between doctors and patients. The Dr.Chat model is specifically designed for the medical domain and has been fine-tuned using real-world patient-physician conversations to provide accurate diagnoses and suggest appropriate treatment options The Dr.Chat backend is built using FastAPI and leverages the state-of-the-art AI21 LLM model, trained on a massive dataset of actual conversations between doctors and patients, consisting of over 50,000 conversations. This large dataset ensures that the AI model has been trained on a wide range of medical topics and conversational styles, allowing it to provide accurate and personalized medical advice to users. Dr.Chat not only provides users with accurate and personalized medical advice but also generates comprehensive medical reports summarizing the conversation with the chatbot. With features such as user authentication, login, and signup, Dr.Chat ensures that users' data is kept secure and private. Additionally, the app provides a user-friendly history of the conversation and summaries generated, helping users keep track of their medical history. Whether you're a healthcare professional or a patient, Dr.Chat is designed to revolutionize the way we interact with medical advice. Join us at the AI hackathon and see how our innovative app can help you make informed decisions about your health!

Emo-Pulse helps to 'listen' to live sales meeting and conversations and transcribes the text from the conversation. It also provies an autocorrect feature for the transcribed text. It can provides the features for sentiment analysis, conversation summary, toxicity analysis ( to check if the conversation is getting toxic) and further suggestions for next dialogues and steps in the sales. We use the generate endpoints of Cohere for sales suggestions and conversation summary (by providing the necessary examples for training) and the classify endpoints of Cohere for sentiment analysis and toxicity analysis.

We bring medically aware AI21 studio's MedBrief, which provides a summary of a lengthy patient report and catches significant medical terminology to save review time by 40%. WE also provide specifics like medicines, lifestyle advises and other medical terminologies based on the report.

🌍 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.

📅 May 11–19, 2026 • May 11 – 19: Online Build Phase. • May 18 (Hybrid Build Day – Online & Onsite) • May 19 (Demos & Awards – Online & Onsite) 🌍 Hybrid event. Join online from anywhere or build onsite at the venue. 🏆 Winners will be announced live on stage at the AI & Big Data Expo North America on May 19 📍 On-site Venue (May 18–19): San Jose McEnery Convention Center, CA, USA 🤝 Join solo or with a team. 🎟 FREE AI & Big Data Expo North America ticket included for participants that are able to join the on-site phase. 🏆Compete for a $10,000 prize pool 🧑‍💻 Apply now to design, prototype, and showcase enterprise-ready innovation.