Muhammad Kashif Abdullah@muhammad_kashif_abdullah935

VISTA is a multimodal video captioning system that generates accurate, context-aware captions by combining visual scene understanding with audio information. Instead of describing an entire video with a single generic caption, VISTA analyzes each scene individually to produce richer and more meaningful descriptions. The system first detects scene boundaries using histogram-based cut detection. For each scene, it extracts a representative key frame for fine visual detail, and a sprite sheet, a grid of frames to capture how the scene changes over time. In parallel, it transcribes spoken dialogue and identifies background sounds such as music or environmental noise, giving the captioning step additional context beyond what is visible. Each scene is described individually by a vision-language model, so its full attention stays on one scene at a time instead of splitting across the whole video at once. When a video has more than one scene, these per-scene descriptions are then combined into a single coherent video-level description. That description is passed to the Gemma model again, along with a detailed persona prompt for each of four tones formal, sarcastic, tech humor, and everyday humor instructing it how each style should sound. VISTA also runs on a fallback chain across three LLM backends, so a failure or empty response from one backend does not stop a caption from being generated.

The Resume Shortlister is an AI-powered system designed to automate and streamline the initial stages of the hiring process. Instead of manually going through large volumes of resumes, recruiters can rely on the system to parse documents, extract key information, and evaluate candidate profiles with high accuracy. By analyzing elements such as skills, experience, education, certifications, achievements, and job-specific keywords, the system builds a comprehensive understanding of each applicant’s qualifications. The core functionality involves converting unstructured resume content—PDFs, DOCX files, or plain text—into structured data suitable for analysis. Once the information is extracted, the system intelligently matches it with the requirements provided in the job description. Factors like required skills, relevant experience, industry background, and preferred qualifications are all considered during this matching process. Based on these evaluations, the system generates relevance scores that reflect how closely each candidate aligns with the role. Using these scores, the Resume Shortlister ranks and shortlists top applicants, allowing hiring teams to quickly identify the most promising candidates without spending hours on manual screening. This automated ranking not only accelerates the hiring process but also ensures fairness and consistency. Unlike human evaluation, which may vary from person to person or be influenced by fatigue, the system applies the same criteria to every resume, making the selection process more reliable.

Managing recurring bills in today’s fast-paced world can be stressful and time-consuming. People often forget due dates, face late fees, or struggle across multiple payment platforms. NexaPay addresses these challenges by combining conversational AI, blockchain automation, and secure digital payments to deliver a truly hands-free bill management experience. Built on the Arc blockchain, NexaPay allows users to interact naturally via chat to check bills, receive smart reminders, and make instant payments in USDC a stable digital currency. Its conversational agent understands context, retrieves bill details, and confirms user intent before executing transactions, ensuring both convenience and security. Through Circle Arc’s developer-controlled wallet sets, each transaction is cryptographically secure, transparent, and auditable. NexaPay also uses intelligent scheduling and alerts, notifying users of upcoming due dates and executing payments automatically upon confirmation. Beyond automation, NexaPay leverages Retrieval-Augmented Generation (RAG) to analyze bill documents via OCR and AI embeddings, enabling precise responses to queries like “What is my total electricity charge for December?” By merging AI reasoning with trustless blockchain payments, NexaPay transforms bill management into a faster, safer, and fully autonomous process, redefining the future of personal finance.

Are you visually impaired? Do you feel left out? Do you find it difficult to navigate? Do you always have to wait for someone to help you? Do you feel sad when you bump into something and people laugh at you? When you’re crossing the road and see a car coming toward you, you can instantly estimate how far away it is. This quick, almost instinctual judgment comes from your spatial understanding of the surroundings. Based on this sense, you can decide whether to speed up, stop, or keep walking. VisioMate, powered by computer vision, works in a similar way to help visually impaired people navigate obstacles. It captures objects in real time and guides users by telling them what is in front of them.

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. ADP members can access AI tutorials, experts, and community support. In addition, new members can claim a free month of DeepLearning.ai pro, $100 in AMD GPU cloud credits, and $50 in Fireworks AI API credits. All hackathon participants will receive on hackathon launch additional compute and API credits for their projects. Ready to build what's next on AMD? Sign up with AMD below to secure your spot.

⏱️ You’ve got 48 hours - create, build, and launch! 🌟 Get direct access to Qubic, mentors, and ecosystem support. 💰 Compete for your share of $44,550 in prizes across two tracks. 🤝 Work solo or team up to design the next wave of blockchain innovation. 🧑‍💻 Sign up and start building on one of the world’s fastest decentralized networks..

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!