@valerioamirani@ZenovayDX Working on a household management website this weekend to keep bills, receipts, warranties and important docs in one place. Might have to try Zenovay to track how people use it :)
@codewithkhalil Stateless. You can use shared session storage like Redis so any server can handle a user's request without relying on local session state.
@bebuu_ai Working on a household management app this weekend. Somewhere to keep all your bills, receipts, warranties and important documents in one place instead of having them scattered everywhere.
What about you?
Seeing a lot of “let’s connect” posts lately, so why not? Drop your website in the replies and I’ll have a look and let you know if I spot any issues or things that could be improved.
kimi k3.1 pro is basically tied with opus 5.5 now. 65.7 vs 66.4 on terminal bench, and it actually beats opus on swe marathon. $2/$10, 1M context, weights go open nov 4. the gap everyone talks about just isn't there anymore
Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model.
🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal.
🔹 Full-project coding: games in Unreal, Unity & Godot.
🔹 Excels at parametric 3D modeling in Blender & CAD.
🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%.
🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks.
Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on https://t.co/85jA71X9gw, Kimi Work, Kimi Code, and the Kimi API.
Open weights by Nov 4, 2026.
🔗 API: https://t.co/XCrgjXAqMw
🔗 Tech blog: https://t.co/5iPP0MBej6
@matsiiako@infisical rotate first, philosophize later. the dynamic secrets point is the one people skip and its the one that actually limits blast radius.
@tinkerlog multimodal embeddings going open-weight is quietly a big deal for RAG. half the retrieval stacks out there are held together by a closed API and hope.
@amanaryan23 youre right, and the irony is the better the models get the more this is true. writing code was the commodity part, judgment was always the job.
@corsaren thats a solid pitch honestly. the rare combo is someone who can do the agent design AND explain it to a client without their eyes glazing over.
@m_zokov I worked on a chatbot for 10 books, each with around 800 pages. Using an agent for every query would be expensive when RAG can retrieve only relevant information.
Yes sub-agents would be able to perform the task but it would still cost a lot don't you think?
@chiefkittenme Hey Kate, I genuinely like what you're building with Tatako. I think a mobile app would be a great addition, especially with offline access so kids can learn even without an internet connection. I'd love to help you build that if you're open to it!