the future of all work is this. You must define: - a goal - the criteria that define it - the verifier that makes sure it is achieved - the sensors that inform the verifier - the actuators that affect the sensors - The envelope that contains the sensors and actuators

updated claude-code to use fable this is the buggies i have seen claude-code is this fable at work?

En Hirevoice buscamos >> Founding Engineer (60k-100k) - Remote o Barcelona [agentic coder] ***pls🙏 repost/max.difusión GRACIAS!!*** Mucha info en... 1/2

lol the claude extension is the wors ask: "please format this document" ai: "Oh perfect, ill write an app script for that, search the strings, and format with functions :D"

Lol for awhile i was looking at this and thinking: ugh when will the opensource models ever get to SOTA, then i remembered that opus 4.7 is like 3 months old and just last week was outdated FUCKING UNBELIEVABLE GREAT WORK

So much better to work with voice ai by having my own workbench than to deal with provider BS Frameworks like livekit are the ones doing most of the work

Been using this internally for a couple weeks, this is the best solution so far

TesterArmy is the simplest way to QA your website or mobile app. It runs real tests across browsers and devices, catches regressions on every PR, generates tests from natural language, and much more. Try now and start testing in minutes: @TesterArmy

RT @SIGKITTEN: oh no what happened to all the principals and morals from last month

holy shit, how did the attackers find a large enough uptime window to get in?

looking at gofeatureflag im becoming go pilled, so cool to have binaries run cheaply and neatly alongside other services given how performant go is im not afraid of running stuff as sidecars

nobody is prepared for how much opus-4.7 costs to produce code, shits crazy

b2b event planning app · components

FireTail landing exploration

Today is a hard day. I shared this note with the @linear team today: We’ve made the difficult decision to increase our workforce. This is not a cost-cutting exercise or a reflection of anyone’s performance. We’re simply reimagining every role for the agentic AI era. We’re hiring. We’re sorry about that.

Pasado mañana Jueves en BCN (7 Mayo) de 10 a 12: ME PONGO PINGANILLO: >>> MUST: Ven sólo si eres reclutador o co-founder reclutando (sino vas a tirar tu tiempo a la basura) 1/3 >>

been doing a lot of work with just text using wisperflow and chatting with the AI. 400k tokens is barbarically big

Extraordinary scenes on my TL Guys there’s more than one way to optimize AI on a task. If you’re working on harnesses try to slowly add all these in your bag. The classic way is to update the weights (RL)… The modern way is to optimize prompts/context (Dspy optimizers/GEPA)… and the hypermodern way is to self evolve the codebase itself (auto research/alpha-evolve/darwin-godel variants) All of them need an eval dataset of prompts/task scenarios, a rubric of success, and an initial forward pass (harness model) to learn. They just update different things to get your system to better evals. There’s nuance to each. There’s a time and place for all of them.

Gotta start wrking on getting familiar with this. Pretty sure this is also how you get better results with cheaper and faster models