Jev unlocked a new way to stop AI agents from spinning.
Agents do not always repeat the same action. Sometimes they try different things based on the same wrong assumption.
I built a semantic circuit breaker around that.
In one stuck run:
12 model calls -> 4
19.9K tokens -> 5.1K
https://t.co/fR44bfNDrA
Some use cases Jev unlocked or rather made it practical at scale:
- Optimized power grids
- Smarter robots
- Autonomous vehicles getting more safe
- Inspecting infrastructure with drones
- Faster scientific discovery
- Social media algorithms more addictive
- Smarter real-time trading
- Better anti-cheat in games/casinos
We were at YC for #vibecon where we built an AI agent that provides predictive cooling for AI data centers.
Traditional cooling is reactive. It waits for GPUs to get hot. We flipped that: our system predicts thermal load and pre-cools before throttling starts.
Everyone is watching reel. The delivery guys are watching reels while i open the door. The delivery partners are watching reels on their bikes. Auto cab drivers are watching reels. Wives are also watching reels. People at the gym are watching reels between sets. Monks are watching reels between meditation. Kids are watching reels at the park. Nannies are watching reels while pushing the stroller.
Is there anyone who is not watching reels?