tanzeel shaikh@tan29

**The Problem** Traditional anti-fraud systems treat every partner as an isolated row. They check thresholds — "Did commission exceed X?" — but affiliate fraud rings operate as coordinated networks: fake clients, mirrored trades, timed deposits. Viewed row-by-row, everything looks normal. In our analysis, 60% of all commissions ($61M out of $101M) flowed to fraudulent partners that rule-based systems would miss. **Our Solution: Deriv Guardian** We shift fraud detection from row-level to graph-level with three layers: **1. The Brain (Kumo.ai Relational Graph Transformer):** We model the affiliate ecosystem as a graph — 2,358 account nodes connected by 6,883 timestamped edges (referrals, trades, commissions). The model performs message passing across the network, learning relationship patterns — not just transaction features. Result: 94% precision, detecting 65 out of 68 fraudulent partners . **2. Temporal Intelligence:** A timeline visualization enables "time travel" through trading data, revealing how fraud rings evolve from grooming (small test trades) to active fraud. The graph transformer detects structural anomalies 2-3 weeks before traditional systems. **3. GenAI Copilot (Azure OpenAI):** Click any flagged partner and AI generates a 3-bullet investigation report — Evidence, Impact, Recommendation — reducing investigation time from hours to 30 seconds. **Data Engineering:** No public affiliate fraud dataset exists, so we built one. From IBM AMLSim (5M transactions, 518K accounts, 370 fraud rings), our Python pipeline assigns Partner/Client roles via network in-degree, maps to Deriv's schema with 7 instruments, and injects realistic fraud: 714 opposite trades (mirrored BUY/SELL within 60s) and 221 coordinated bonus abuse deposits. **Tech:** Kumo.ai (Graph Transformer), Azure OpenAI GPT-4o, FastAPI, React, Python/Pandas,Docker **Impact:** $61M in preventable fraud identified. Early detection during grooming phase. Investigation time: hours → 30 seconds.

Supply Chain Quality Control Automation - Business Case The Problem: Manual quality control in supply chain operations is slow, error-prone, and costly. Wrong items get shipped, damaged goods slip through, and warehouse staff spend hours on repetitive inspection tasks. The Solution: An automated quality control system that uses AI to inspect boxes and verify labels, catching errors before they reach customers. Problem 1: Wrong Items Get Shipped: What Happens: Package gets shipped to customer Customer receives wrong item Customer complains and returns item Company pays for return shipping ($25-50) Item needs to be restocked Correct item needs to be shipped again Customer satisfaction drops Manual visual check by worker Supervisor spot-checks 10% of packages Most errors caught only after customer complaint Problem 2: Damaged Goods Enter the Warehouse: What Happens: Damaged box enters warehouse Gets stored with good inventory Eventually gets shipped to customer Customer receives damaged goods Customer rejects shipment Company must replace item Original damaged item is now unsellable Receiving clerk inspects 5-10 boxes per shipment Takes 15-20 minutes per shipment Human fatigue causes missed defects No standardized criteria for "damaged" How It Works (In Simple Terms) Imagine you have a supervisor that never gets tired, never misses details, and works 24/7. That's what this system does. The System Has 4 "Specialists": The Box Inspector - Looks at every box and checks for damage The Label Checker - Reads labels and verifies they match the product The Order Validator - Checks if the order is correct in the system The Problem Solver - Handles issues and alerts supervisors How They Work Together: When a box arrives, the Box Inspector checks it automatically When a package is repacked, the Label Checker verifies the label matches If something's wrong, the Problem Solver stops the process and alerts a human Everything happens in seconds, not minutes.

Actify is an AI-powered multimodal assistant powered by Gemini models and Google AI Studio , designed for people who constantly save reels but never revisit them. Instead of leaving your saved content in chaos, Actify automatically analyzes every reel you save and turns it into a beautiful, structured, enriched, and recallable memory and extends action onto it. **How It Works** 1. The user saves or uploads a reel (Instagram, TikTok, YouTube Short). 2. Actify downloads the video, extracts frames, and extracts audio. Gemini 1.5 Pro / Flash performs multimodal reasoning to classify the reel type and generate structured data. The result is stored inside Supermemory.ai for long-term memory and semantic search. 3. The user searches their saved reels using Actify’s AI-powered search. Examples of queries: - “Show me my 10-min HIIT workouts.” - “Recover the pasta recipe from that reel.” - “Show travel places from reels I saved last month.” 4. The structured data is reconstructed using Gemini Pro models and formatted into richer, more actionable context. 5. Actify then enhances and enriches the output with actions based on the reel type, such as: - Google Maps links - Product shopping links - Recipes with ingredients and steps - Workout timers and round structure Don’t just save — Act on it!

Join Deriv and lablab.ai for a high-intensity hybrid hackathon where top builders create AI prototypes, demo their work, and get fast-tracked to interviews. When • Sat, Jan 31, 2026- On-Site Pre-Hackathon: Deriv AI Meetup (Dubai) • Fri, Feb 6 - Opening words, Challenge, and Hackathon kick-off. (Dubai) • Sat, Feb 7 - Build and ship your prototype - 12 hours (Onsite +Online) • Sun, Feb 8 - Demos, judging, and next steps for standout teams. (Onsite +Online) On-Site Pre-Hackathon: Deriv AI Meetup Sat, Jan 31, 2026 · 2:00–6:00 PM (GST) · Dubai In-person talks, workshops, team formation, and networking with the Deriv team. 👉 Deriv AI Meetup registration What's in it for you • USD $10,000+ in prizes across categories • Mentors available to unblock you when you're stuck • Fast-track interviews for top performers - skip the queue • Real problems to solve - not toy demos Format • Teams of 1-3 • Hybrid: Dubai on-site or fully remote • On-site spots confirmed after screening Location • One by Omniyat, 24th Floor, Al Mustaqbal Street, Business Bay, Dubai, UAE • Travel and accommodation not provided. Apply now • Spots are limited. Shortlisting begins right after the event.

Leverage Orchestrate’s digital skills to design solutions that automate repetitive tasks, orchestrate workflows across tools, and empower employees to focus on high-value work. ⏳ Complete your project in just 48 hours and showcase your innovation. 🚀 Gain hands-on experience building with IBM watsonx Orchestrate. 💡 Develop practical AI agents that reshape the way teams and organizations operate. ⭐ Learn directly from experts who will support you throughout the event. 🤝 Team up or work independently to bring your vision to life. 💰 Prizes: $10,000 🧑🏻‍💻 Sign up before the Kick-Off Stream and start shaping the future with IBM watsonx Orchestrate!

Join the next edition of our flagship AI hackathon — now featuring Google, AppliedAI, and Qdrant as official partners, with over $45,500 in prizes and exclusive partner challenges. After an incredible launch earlier this year, we’re back with the next edition of our AI hackathon series leading up to the /function1 AI Conference & Exhibition in Dubai. Tap into the power of AI to innovate, build, and take your AI skills to the next level in this high-impact hybrid experience. 📅November 14–19, 2025 • Nov 14–18: Collaborate and build online with AI enthusiasts from around the world. All projects must be submitted by end of day on November 18 • Nov 18 (Dubai): An exclusive on-site build day at Festival Arena Dubai for selected participants. • Nov 19 (Dubai): Live on-stage pitching & winners announcement at the /function1 Conference. 🌟 Get support from expert mentors throughout. 👥 Go solo or team up. 📍 Please note: On-site participation is by invitation only. Travel and accommodation will not be covered for selected participants. Further details will be shared with those accepted. 🧑🏻‍💻 Secure your spot now—sign up before the Kick-Off Stream!