Wild year after leaving Wix! Learned a ton about scaling AI products, now building something new at the intersection of business apps and AI (still in stealth ๐คซ)
I'll be at #dreamforce2025 in SF next week. Down to meet if you want to talk:
- Enterprise AI
- Business platforms
- What we're cooking up
Shoot me a DM ๐
@salesforce
But with the right context, tools, evaluation, and reflection, they can achieve goals that no standalone planner can.
If youโre diving into actionable AI with agents, this chapter by @chipro is a must-read. Itโs a great resource that captures the current state of agent frameworks and how theyโre evolving.
https://t.co/SJ62HWlFqP
#AI #LLM #Agents #ActionableAI
Auto-Regressive LLMs can't plan
(and can't really reason).
"While our own limited experiments didn't show any significant improvement [in planning abilities] through fine tuning, it is possible that with even more fine tuning data and effort, the empirical performance may well improve. But all that such fine tuning is doing is converting the planning task into a memory-based (approximate) retrieval. It doesn't prove that LLMs are able to plan."