Kaggle

📢 The 5-Day AI Agents: Intensive Vibe Coding Course With Google is now available as a Kaggle Learn Guide! Missed the live sessions from June 15 to 19? Access whitepapers, podcasts, codelabs, and recorded livestreams at your own pace.

The Biohub - Cell Tracking During Development competition, hosted by Biohub, is live! Build machine learning models to identify cell divisions, track changes over time, and reconstruct lineages in real 3D data. Your work will help solve a huge challenge in biological research.

The Pokémon Trading Card Game AI Battle Challenge is now live - in partnership with @Pokemon_cojp 📢 Build AI Training Agents through two connected competitions focused on strategic gameplay in the Pokémon Trading Card Game environment.

Develop systems that can adapt to complex board states, evolving battle conditions, and diverse opponent strategies. Prize pool: $240,000 (awarded through the strategy category track) Simulation entry deadline: August 9, 2026 Strategy Category entry deadline: September 6, 2026

The AI Agent Security - Multi-Step Tool Attacks simulation is now live! In partnership with OpenAI, Google and IEEE, your challenge is to build an attack algorithm that stress-tests tool-using AI agents in a deterministic offline benchmark.

1H-VideoQA is now available on Kaggle Benchmarks! Developed by Google DeepMind back in 2024 (Antoine Yang) and now updated with latest SOTA models, 1H-VideoQA is a 101-prompt benchmark for long-context video comprehension and temporal episodic reasoning across hour-long YouTube footage.

Top of the leaderboard: 🥇Gemini 3.5 Flash: 80.2% 🥈Gemini 3 Flash Preview: 79.2% 🥉Gemini 2.5 Pro: 78.9% Models must process raw frames as native tokens to locate seconds-long events hidden in an hour of video; accuracy scales logarithmically with frame density.

Show us how you'd take an idea and turn it into a working benchmark. We're picking 5 submissions to win exclusive swag and a social shoutout.

How to enter: 1️⃣ Build a task locally with the write-kaggle-benchmarks skill 2️⃣ Push it to Kaggle Benchmarks and run it 3️⃣ Post your Task link plus a screenshot or video of how you did it Must tag @kaggle. Deadline: July 1.

Earlier today we released local development for Kaggle Benchmarks. 🚀 You can now write, validate and run AI evaluation tasks directly from your preferred dev environment — VSCode, Antigravity, Claude Code, and more. Go from idea to working eval with the write-kaggle-benchmarks skill.

Watch the full demo here 👇 youtu.be/c7B8vyehyUA

Join the 5-Day AI Agents: Intensive Vibe Coding Course with Google happening June 15–19. Learn from Google experts through theory sessions, hands-on labs, and a capstone challenge alongside the Kaggle community.

Get ready - tomorrow June 3 at 9:30 AM PT / 12:30 PM ET, the 5-Day AI Agents: Intensive Vibe Coding team is going live on YouTube with Google Cloud!

Course creator Anant Nawalgaria, cohost Smitha Kolan and course leader Brenda Flynn from the Kaggle team, give a sneak peek at the curriculum, the history of the course and what to expect over the 5 days- then take your questions live.

We just opened the collective ML expertise of Kaggle’s community – discussions, solution writeups, debugging threads – to your coding agents. The new Kaggle CLI (v2.2.0) makes this knowledge your agent’s knowledge.

This upgrade also ships OAuth authentication (kaggle auth login), Benchmarks CLI, JSON output for scripting, and rate-limit handling. Upgrade now with pip install --upgrade kaggle ✨

When an AI agent succeeds, was it the model or the skill it was given? Launching today with BenchFlow — the BenchFlow AI Agent Skills Community Hackathon. Build skills that lift agent capability without crossing safety boundaries.

Join the 5-Day AI Agents: Intensive Vibe Coding Course with @Google! Learn from Google experts through theory sessions, hands-on labs, a capstone challenge, and connect with a vibrant community!

We’ve just added Gemini 3.5 Flash to Kaggle Benchmarks! You can now stress test this new model using your own custom evaluation tasks and benchmarks at no cost. 🧵

The real benefit here is running side-by-side comparisons. Benchmark Gemini 3.5 Flash against leading models using a single, unified API — no more switching between different providers to see which model actually wins on your specific task. Get started: www.kaggle.com/benchmarks

Can your AI agent win our new simulated challenge? 🧑‍🌾 The all-new capstone challenge for our 5-Day AI Agents: Intensive Vibecoding Course with Google is here: Kaggriculture!

What if you could map the subsurface without ever seeing it? 🌍 ROGII Wellbore Geology Prediction is now live on Kaggle! Your goal is to build ML models to predict geology along horizontal wellbores and improve drilling accuracy.

Registration is now open for the 5-Day AI Agents: Intensive Vibecoding Course with Google 🚀 This no-cost course is designed to help builders learn how to design, build, and use AI agents using the latest concepts, technologies and skills.

What does it take to build an AI agent that can compete in a multi-agent environment? Inspired by the 2010 Planet Wars, our newest simulation challenge tasks you with building agents for 1v1 and 4-player FFA matches. $50,000 Prize Pool Entry Deadline: June 16, 2026

Calling NeurIPS 2026 researchers! 🚀 The CFPs for the evaluations & datasets track are officially out, and we want to support you if you're working on a submission. When you build your benchmark on Kaggle Benchmarks, you get complimentary compute, top model access, and exposure to 30M+ users.

Just released! The Neurosynthetic Research Institute is launching the IJCAI-ECAI 2026 NeuroGolf Challenge. Implement ARC-AGI transformations by designing minimal neural networks. How simple can a model be and still solve the task?

You’ll work with tasks from the public training set (v1) to submit ONNX models that are both accurate and as small as possible. Solutions will help define how much computation these tasks truly require. 💰 $50,000 Prize Pool ⏰ Entry Deadline: July 8, 2026