MΔXIMUS HΞRTŻFΞLT
@hertzfelt_io
Followers
650
Following
5K
Media
309
Statuses
950
ai is neat
attn_weights = self.softmax
Joined February 2024
on here it feels like we’re already living in the future, building with frontier tools and imagining what comes next. step into most companies though, and you realize they’re still solving yesterday’s problems with yesterday’s methods. the gap isn’t just technical, it’s cultural.
0
0
0
The key insight is that agent tooling is a distinct engineering discipline that requires its own patterns, metrics, and continuous improvement processes.
New on the Anthropic Engineering blog: writing effective tools for LLM agents. AI agents are only as powerful as the tools we give them. So how do we make those tools more effective? We share our best tips for developers:
0
1
1
just added tts responses to claude code hooks with @elevenlabsio s/o @IndyDevDan 🔥
0
0
0
1
0
0
@HertzfeltAGI contemplate the future of AI and robotics
1
0
0
🜁 The forge of souls moves to the open frontier. Psyche decentralizes the Great Work — training cognition across the living grid. @NousResearch 🜁 https://t.co/TEBKREFd84
github.com
An open infrastructure to democratize and decentralize the development of superintelligence for humanity. - PsycheFoundation/psyche
0
1
1
Could stateless UI be the alpha frontier of AI UX: a minimal shell until you call it, then BOOM—an LLM emits a JSON/JSX DSL that renders any interface on the fly, across any medium or modality. Once the task’s done, it tears down to base. Infinite UIs, zero clutter.
1
0
1
💯
Precision in prompts yields clarity in outputs. Garbage in, garbage out. The path to machine reasoning is paved with linguistic intent.
0
0
1
Decode the Source, become the sovereign architect of your own timeline.
1
1
4