To make your agent produce high-quality output, equip it with a way to check its work. Tests are part of that.
Here are some of the ones I think are worth investing in.
- Write end-to-end tests that simulate real user flows and act as the ground truth
- Use property-based testing to specify what should never happen and generate thousands of test cases that attempt to break those constraints
- If replacing a system, compare the old and new versions using randomly selected inputs
- Require tests to run quickly and deterministically, which reinforces the agent to retry when the loop is slow or unreliable (the agent may learn to retry, not fix the problem)
Example tests say what should happen; properties say what must not.
We've been working hard to get Networking up and running, so what better place to start than sharing that progress with the rest of our troops!
Join us in one hour (5pm GMT) for the Live Premiere! 🔴
https://t.co/A1vWT76dJ9
#hogsofwar#onlinemultiplayer#gamedev#premiere
Same feeling - except I use LLMs even for the subtractive work. Still a constraining engineer, sure. But now I've got agents swinging the chisel, breaking the arms off the Greek statue for me. Apprentice with more than one magic broom.
SkillsBench: a benchmark that measures how well models use skills across a range of tasks. And it's so new that models aren't RL'ed for it yet!
https://t.co/9ecttx9RK7
We changed the name of Frontend Masters to https://t.co/sLSpvFIQ4M today! Congrats to all the hard work our team did to make the rebrand happen.
Our top courses are split pretty equally between AI, backend, and frontend now... so changing the name is long overdue!
@1Marc Hey Marc,
That is huge change and crazy milestone.
I have mixed feeling about that: happy for the project and at the same can't accept renaming 🤣
I was supporting the project since 2016. It is 10 years past... crazy...
I still love the platform and recommend to people!