We're excited to announce our newest challenge with Bright Data!
Running through May 18 May 25, the Bright Data Real-Time AI Agents Challenge invites you to build intelligent AI agents and systems that can autonomously interact with the web, retrieve live data, and make decisions based on the most current information available.
Whether you're an AI engineer, data scientist, or vibe coder, this hackathon is the perfect opportunity to build something awesome.
One talented developer will take home $3,000, an exclusive DEV badge, and a DEV++ membership! All participants will receive a completion badge.
We hope you give it a try!
Our Prompt
Your mandate is to leverage Bright Data and build an AI agent or system powered by real-time web data.
The most powerful submissions will utilize Bright Data's MCP server to enable all four key actions:
- Discover: Find relevant content across the open web
- Access: Navigate even the most complex or protected websites
- Extract: Pull structured, real-time data at scale
- Interact: Engage with dynamic, JavaScript-rendered pages as a human would
The MCP server is specifically designed to empower AI agents to perform these actions seamlessly, and we encourage you to incorporate all four capabilities in your project.
The most important aspect? Your solution should showcase how access to reliable web data improves AI performance in solving real-world problems.
Recommended Tech Stack 🛠️
Use the latest open-source AI tools to build your project. Here’s a modern stack to get you started (but feel free to get creative!):
- Frontend: Next.js, Vercel, Streamlit
- Embeddings & RAG Libraries: Nomic, Cognita, LLMWare, JinaAI
- Backend & Model Access: LangChain, Netflix Metaflow, Hugging Face, FastAPI, Ollama
- Data & Retrieval: Postgres, Milvus, Weaviate, PGVector, FAISS
- Large Language Models: LLaMA 3.3, Mistral, Gemma 2, Qwen, Phi
You can get to know the Bright Data platform by utilizing their docs and tutorials:
Project Ideas 💡
- Real-time news or sentiment trackers
- Market or price monitoring bots
- LLMs with live data pieplines
- Multi-agent research assistants
- Crawlers for data extraction from forums like reddit
- Or anything else you can imagine!
Important Dates
- May 7: Bright Data AI Web Access Hackathon begins!
May 18May 25: Submissions due at 11:59 PM PDT-
May 29June 5: Winners Announced
We can't wait to meet the agents you build! Questions about the challenge? Ask them below.
Good luck and happy coding!
Good luck everyone!