Use local rules to sort waste from photos
Have Gemini check local recycling rules online, identify items in a photo, and assign each one to compost, recycling, or trash.
- What they did
- A developer used Gemini Robotics-ER 1.5 as the reasoning layer for a robot-style sorting task. The model searched the web for local waste guidance, recognized the objects in view, and mapped each item to a bin before passing the plan to a vision-language-action model or custom functions that would move the objects. In the demo, it also explained its choices and pointed at each item.
- How it went
- The post showed a rendered animation of a spatial sorting plan with text and pointers mixed in. It did not report end-to-end sorting accuracy, and the model was still in preview.
- Tips
- Treat the model’s safety features as only one layer of protection. Real deployments still need emergency stop controls, collision avoidance, and a fresh risk review.
Try it yourself with Gemini
You are helping sort items for disposal. First, look up the current recycling and compost rules for [city or waste provider] and quote the source page for each rule you use. Then review the items in [photo folder or camera feed], identify each object, and assign it to compost, recycling, or trash with a brief reason. If any item is unclear, mark it as unsure and ask me before making a final decision. Before any item is physically moved, send me the full plan for approval and wait for my confirmation.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Have Gemini write robot control plans from vision and speech
A developer used Gemini to turn camera input and robot tools into Python plans, then drove robot actions with voice commands.
Run Spot step by step with Gemini
A developer paired Gemini Robotics ER 2 with Boston Dynamics Spot so it could plan, check, and continue multi-step chores.
Scan open-source projects for possible security bugs
Big Sleep found and reported 20 previously unknown issues in widely used open-source projects.
Test a math agent on hard open questions
DeepMind’s Aletheia, built on Gemini Deep Think, found proofs for four Erdős problems in a large benchmark run.
Review recent patches for nearby SQLite flaws
A Gemini agent followed one fixed SQLite bug into surrounding commits and found another memory-safety issue before it shipped.
Search for room-temperature magnetic semiconductor candidates
A Claude Opus 5.5 agent team screened Luttinger-compensated magnets and surfaced two predicted room-temperature magnetic semiconductor candidates.