When I lost my job at @HelixeSportsFX I thought I’d never find a job that I loved as much
Over the last 8 months I became obsessed with building with AI.
Now I’m working full time on an engineering team developing an AI platform.
Work harder, nobody cares.
One prompt that has improved my workflow:
“Spawn an independent senior engineer subagent that continuously audits your work while you execute the plan.
Have it review completed tasks for:
bugs
flawed assumptions
deviations from the plan
missed edge cases
@thsottiaux they legit reset for us an hour before you did. Yet again another reset on the same day that you reset last week at the same time. This reset probably benefit 3-5% of users max.
Resetting everyone’s usage the day 95% of the userbase already resets to celebrate 5 million is a massive L. Could wait til Monday or Tuesday. @thsottiaux
@YoavCodes@lalathalala_ Okay I can understand that is a lot. IMO should be a phased depreciation once rust is a few versions in and tested. The internal evals, testing + mythos thrown at it must make them feel good. I doubt Anthropic would ok if there was concern.
@YoavCodes@lalathalala_ For anything that you deem production dangerous should still be hand reviewed but you can 100% have 5.5 copy word for word, write a script and verify 100% parity consistently.
@YoavCodes@lalathalala_ It’s just not true that you can’t successfully have a model extract text line for line and be 100% parity. I didn’t mean write a skill, I meant it comes down to the expectations and context of the stated goal and proof of verification.