Divya
203 posts
Divya
@divzzrk
Lead GenAI Engineer @Hashedin | Ex @UnicourtInc | Building AI agents and workflows that do my work so I can doomscroll in peace 🎧
Katılım Kasım 2019
205 Takip Edilen40 Takipçiler
This is crazy !!
Cursor now has built a Kanban board where you can just drop in tasks and the agent will pick those up and complete them.
English
What if every open issue on linear is picked up by one agent and it solves it and humans are only for review? That’s exactly what symphony solves
Checkout the latest post:
openai.com/index/open-sou…
English
Divya retweetledi
Happy Friday!
We just put DeepSeek-V4-Pro up on build.nvidia.com. It’s the world’s largest open source model at 1.6T parameters, and you can run it for free running on NVIDIA Blackwell GPUs.
Try the NVIDIA NIM API → build.nvidia.com/deepseek-ai/de…
English
@piyush_lakheda Having a persistent compounding artifact. Checkout LLM Wiki by Andrej Karpathy
English
Context Memory, the biggest challenge? Isn't it in Agentic Ai? How are you guys solving it? Except the Summary concept! Drop in! Let's connect if you have really solved it! Would like to discuss
English
How often will you post, if in case X comes up with a strategy of having a queue mechanism for filtering duplicate posts?
English
Not Vibe Coding!
Checkout, What I do as Agentic AI Developer/GenAI Developer in 2026?
youtu.be/ABzfQZi3IKs?si…
via @YouTube
YouTube
English
Have you also experienced decrease in model performance before opus 4.7 dropped in?
ClaudeDevs@ClaudeDevs
Thanks to the entire Claude community for giving feedback and continuing to build with us. Read the full post-mortem here: anthropic.com/engineering/ap…
English
Divya retweetledi
New in Claude Code: /ultrareview (research preview) runs a fleet of bug-hunting agents in the cloud.
Findings land in the CLI or Desktop automatically. Run it before merging critical changes—auth, data migrations, etc.
Pro and Max users get 3 free reviews through 5/5.
English
Divya retweetledi
Codex is introducing true long-term workflow orchestration.
With the addition of scheduled automations, thread reuse, and a new persistent memory system, the agent can retain project conventions, go to sleep, and autonomously "wake itself up" later to keep pushing a long-running task forward.
Andrew Ambrosino@ajambrosino
The biggest shift: Codex can now use your computer alongside you. With background computer use, multiple agents can operate desktop apps in parallel while you keep doing other work.
English
@Amank1412 Codex is currently coming up with better features as well as it just brought in agent sandbox. so why not
English