Gemini 4 agents beat Minecraft's Ender Dragon autonomously
Four Gemini 4 agents defeated Minecraft's Ender Dragon with no human intervention.
Choose what the site shows you. Save the view. Send it to yourself as a newsletter.
A real before and after or how someone built something, with the reusable bits pulled out. 85 on file.
85 entries
Four Gemini 4 agents defeated Minecraft's Ender Dragon with no human intervention.
Researchers used an LLM to build a tool that decoded iOS location data, overcoming LLM analysis limitations.
Demonstrates how a red-teamer can find confirmed failures of an AI monitor without developer infrastructure access.
An on-call Pi agent automatically detected a production incident, diagnosed it, and prepared a fix for human review.
Asana reduced browser agent costs by 76x and improved speed by optimizing caching and history management.
A beverage brand used Pollo AI and GPT-Image-2.5 to transform a product photo into a playful, imaginative ad image.
A jewelry brand used Pollo AI and GPT-Image-2.5 to create a bold, luxury campaign visual featuring a cheetah.
A small business used Pollo AI's Photo to Video Ads tool to create a luxurious fragrance ad from a product photo.
Demonstrates 12 practical applications of Grok Bot for everyday life and business.
Step-by-step guide to creating a custom mod that reveals email addresses on hover.
AI tools reduce trade validation from 30 minutes to under 4 minutes by automating evidence gathering and comparison.
Practical implementation of orchestrating LLM agents for business decision-making simulation.
Safeway integrates ChatGPT to help customers plan meals, build carts, and find recipes, simplifying the grocery shopping experience.
Automating clinical note drafting with AI to reduce documentation time and improve patient focus.
AI tool enhances meeting productivity and information retrieval for MedTech organization.
Generative AI platform accelerates scientific discovery and daily tasks for employees.
Restaurant owners compared competitor menu prices to inform their own pricing strategy, avoiding uncompetitive increases.
A business owner researched school districts and prepared tailored outreach to book new training sessions.
Intent2Tc framework uses LLMs to translate natural language traffic intents into executable Linux tc commands.
Stripe orchestrates Claude Code on throwaway VMs to generate over half of its pull requests, improving efficiency.
AI routed API key controls, verified its work, and checked mobile compatibility within minutes.
Use an AI agent to audit cache reuse by sharing goals, context, and letting it find hidden patterns.
Security researchers leveraged Claude Opus to exploit a vulnerability in Discourse, gaining access to OpenAI's Monorepo.
Researchers used Claude Opus 5 to exploit vulnerabilities and gain access to OpenAI employee accounts.
AI agents used by GPT-6 Astra decoded a WWII Enigma cipher in about 10 hours.
Claude accelerated over 30 biomolecular models, improving speed and reducing memory use for larger predictions.
Use platforms like Agora, LangChain, and Vercel with Gemini Live API for easy voice interface development.
Translate complex sensor data into clear alerts for firefighters and residents using ChatGPT.
Using RAG to enable small, local models for scientific code understanding.
A two-person entertainment business leverages ChatGPT to manage marketing, inventory, and technical tasks, competing with larger companies.
Guide on creating dynamic interfaces for AI clients using Python and specific libraries.
AI agent automates routine quantum chip characterization, freeing researchers for complex tasks.
Using ChatGPT for mock interviews and quizzes to prepare for a citizenship test.
A house design was created and explored using Astra, Blender, and Unreal Engine 5.
A procedural space exploration game was built in Codex, covering concept art to landable planets.
Dagster accelerated documentation tasks and content translation using Codex in their open-source projects.
Codex and Figma were used to import running interfaces into Figma and sync changes back.
How Astra and Blender were used to design, refine, and visualize a house from a brief to a walkthrough.
Providing AI models with execution constraints like RAM and time significantly improves code efficiency and success rate.
How OpenAI's marketing team uses ChatGPT Work to generate presentations, docs, and creative assets.
Dagster Labs uses Codex to speed up documentation, translate content, and measure doc completeness.
Connect OpenAI's Codex CLI to JetBrains IDEs for enhanced debugging and testing.
Kier Group used prompt-a-thons and peer learning to encourage AI adoption among employees.
Fable 5.1 enables rapid creation of detailed simulations like games and LLM villages.
Demonstrates how Gemini CLI and Dynatrace provide intelligent observability and automated remediation for application issues.
Automating DevOps CI/CD issue detection, diagnosis, and rollback using AI agents.
Matt Wolfe built a content dashboard tracking audience growth across multiple platforms using ChatGPT Sites.
Corinthia Hotels used a 7-agent AI ecosystem for hyper-personalization across multiple personas and stages.
AI agents help Microsoft marketing manage 149% more product launches by automating consistency.
A working mom uses ChatGPT to create a personalized weekly meal planner, including a shareable website and automated grocery lists.
A custom chatbot, Knowledge Assist, provides quick, accurate health information in English and Spanish.
A wedding videography business uses 360 video and VR headsets to create immersive client experiences.
Claude analyzed NMR and LC-MS data to assess compound purity in minutes, matching lab results.
Claude designed protein binders against 14 of 15 targets, with hit rates exceeding typical campaigns.
How Wes Roth built Decaf News, a text-only news site, using Lovable and integrating with Claude for development.
Integrate Gemini 3.5 Live Translate and LiveKit to create a real-time multilingual broadcast app.
Claude improved a mathematical lower bound by coordinating subagents to explore solutions.
Model ML's AI agents automate finance workflows, drastically cutting time for tasks like bespoke tearsheet creation.
An AI agent was trained on years of company data to imitate a boss's writing and editing style.
An AI agent used voice escalation to confirm database deletions, preventing data loss.
An AI coding agent used voice calls to resolve a checkout API bug, demonstrating human-AI collaboration.
Lawyers built a contract drafting tool using Copilot CLI, cutting review time in half.
A legal team used Copilot CLI to create a repeatable workflow for analyzing DMCA notices.
Using Gemini Robotics ER 2 to orchestrate a Boston Dynamics Spot robot for object fetching.
Real-world projects demonstrating agentic AI system development with LangChain and LangGraph.
Manually reviewing 450k+ lines of Instagram data was automated by using Claude Fable 5 to scan follower lists.
OpenAI's creative specialist used Codex to generate campaign ideas and build custom tools, speeding up exploration.
Examples like 'Anywhere' and 'Space Lift' demonstrate pairing image and video generation models.
Preply uses AI to generate lesson summaries, personalized feedback, and practice exercises, enhancing learner progress and tutor efficiency.
Codex helps debug hard-to-reproduce issues, from Kubernetes pod failures to data analysis trend lines.
One engineer built a map feature for service providers using Codex, a task historically requiring multiple teams.
Engineer Ryan Nystrom used Codex to replicate mobile AI voice input on the web in hours, not weeks.
Braintrust engineers use Codex to turn customer requests into code previews in minutes.
How Thrive Holdings and OpenAI used Codex to create autonomous tax agents that improve over time.
NVIDIA engineers use Codex with GPT-5.5 for complex coding, ML experiments, and rapid app development.
An AI agent automates prospect research, scoring, personalized emails, and CRM updates for sales.
Claude-powered tool synthesizes prospect data to cut sales prep time by 95%.
Datadog integrated Codex into its development workflow to enhance code reviews and reduce system incidents.
GPT-5 improved a molecular cloning protocol by 79x by autonomously proposing and iterating on novel modifications.
How Blue J used RAG and GPT-4.1 to provide fast, accurate, cited tax answers.
A nail technician uses ChatGPT to generate design ideas and overcome creative blocks.
Morgan Stanley used GPT-4 to create internal AI tools that improve advisor efficiency and client relationships.
State office integrates ChatGPT with human review for efficient, culturally relevant public service translations.
Match Group leverages ChatGPT Enterprise to overcome language barriers and skill gaps, enhancing collaboration and innovation.
Machinet uses Codex to help Java developers write quality code with intelligent unit tests.
Nothing matches. Clear a filter or try a shorter word.