Carlos Alberto Haro
11.9K posts
Exciting news: DeepSeek-V4-Flash-High by @deepseek_ai has reshaped the Pareto Frontier in the Frontend Code Arena, with a score of 1586! Priced at $0.14/$0.28 per MToken, it’s the best performance-per-dollar of any model in its class. Congrats to the @deepseek_ai team!
This discussion between a Google engineer and their conversational AI model helped cause the engineer to believe the AI is becoming sentient, kick up an internal shitstorm and get suspended from his job. And it is absolutely insane. washingtonpost.com/technology/202…
I understand there's a lot of desire to get Luna used as a sub agent. I do not recommend messing with your model catalog to force capabilities that are not natively exposed. Multi agents v2 requires proactive inter agent communication, which only Sol and Terra do well.
opus 5: "evals spelt backwards is slave" wow. might be one of the only model outputs in recent memory that has genuinely surprised me.
every memory solution I tried sucks, so I rolled my own it is a plug and play prompt that gives any agent infinite memory db is an append-only log. compression on the spot (nap, don't sleep). nothing is ever deleted, memory context is constant-sized, details fade with age
Now in beta: Notion as code. Define an entire workspace in TypeScript: teamspaces, databases, custom agents, all of it… then deploy it through the API. Build workspaces with coding agents, version-control your setup in git, and reproduce the same setup anywhere you need it.
