Muhammad Ammar Kaboolio@r6492

This project is an automated video captioning agent designed for the AMD Hackathon. The pipeline processes video content by separating the task into four straightforward steps: Ingestion: The system samples frames from the video and transcribes the audio track into text using a local transcription model. Extraction: The sampled frames and audio text are sent to a Vision-Language Model, which identifies the raw, objective facts of the scene (e.g., physical objects, setting, actions, and lighting). Compression: The raw facts are grouped and simplified into a clean timeline of events, removing redundant information. Synthesis: Finally, the timeline and key frames are fed into a large language model. The model is given strict persona instructions to apply the requested styles (Formal, Sarcastic, etc.) while staying firmly grounded in the factual timeline, ensuring the captions are both stylized and visually accurate.

Scrappy is poised to transform how the world extracts and leverages data from the web. Our journey starts with the launch of the core autonomous API generator, giving anyone—from analysts to entrepreneurs—the power to turn static websites into usable APIs in seconds. This foundation eliminates the fragility of traditional scraping and opens access to web data for a much wider audience. The next step is expansion. Scrappy will move beyond static sites to support dynamic, JavaScript-driven platforms, ensuring compatibility with the modern web. Alongside this, we will integrate multi-agent intelligence, allowing Scrappy to analyze, generate, test, and refine APIs in a collaborative and autonomous way. This evolution will create a more adaptive, resilient, and intelligent data-extraction system. As Scrappy matures, we will enhance intelligence and reliability through advanced monitoring, analytics, and enterprise-ready features. Businesses will not just receive an API—they will gain intelligent data pipelines that monitor themselves, adapt to change, and deliver consistent results over time. This will position Scrappy as both a developer-friendly tool and a robust enterprise platform. Ultimately, Scrappy is about more than scraping—it is about building the autonomous data layer of the future. A world where any website can be instantly transformed into an accessible, structured API. By lowering barriers to access, Scrappy democratizes web data, empowering individuals and organizations to innovate faster, make smarter decisions, and unlock new opportunities. The vision is bold but simple: launch strong, expand capabilities, enhance intelligence, and pave the way toward a web where information flows freely as APIs. Scrappy is not just solving a problem—it’s shaping the future of data accessibility.

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.

🕒 6 days to explore the full potential of GPT-5 💡 Build next-gen AI apps 🤝 Compete solo or form a dream team 🚀 Connect with global innovators 🧑🏻‍💻 Sign up before the Kick-Off Stream to secure your spot!

🕒 7 days to be among the first to experiment and build with OpenAI o1 💡 Build innovative applications utilizing o1's advanced features 🤝 Compete solo or collaborate with teammates 🚀 Dive into a 1-week virtual coding marathon, connecting with industry leaders 🏆 A chance to join the Lablab NEXT acceleration program and fast-track your innovation