developer's journal

Documenting the journey of building software, AI tools, and 3D worlds. Sharing raw progress, technical insights, and design logic

[developer's journal @DevJournaI]

What happens when AI stops writing and starts deciding?

4 days after Jev’s release:
→ Browser agents
→ SEO automation
→ Email routing
→ Context compression
→ Local implementations

We tracked what developers built around it.

Full breakdown ↓ https://t.co/xh6sowThak

[developer's journal @DevJournaI]

Jev is an “internet moment” for the AI industry.

It tells your agents and LLMs what to do next: in milliseconds and almost for free. If set up correctly, you get the 2028 AI engineer’s stack.

The article contains a complete 10-step roadmap: how to build and use a “new brain” for AI from scratch. From installation to real automation use cases.

And the key idea is the Jevons paradox (it’s in the image): an increase in the efficiency of resource use does not reduce its overall consumption—on the contrary, it increases it. That is exactly why Jev does not replace LLMs, but makes them cheaper, faster, and more practical.

📖 [https://t.co/IwrlzomvE6]

#AI #Jev #JevonsParadox #agents #automation

[codila @0xCodila]

Jev is the "Internet" moment for the AI industry

It tells your agents and LLMs what to do next, in milliseconds and at almost zero cost

If you set it up correctly, you will have the AI engineer’s stack for 2028

In this article, I show you how https://t.co/x3qn41ejnr

[developer's journal @DevJournaI]

“Jev is not a chatbot. It doesn’t write text—it makes decisions.”

The author broke down 30 real-world examples of using TypeSafe AI’s new model, Jev: from controlling Mario and Doom to automating business processes.

Key points from the article:
🔹 Jev doesn’t return an answer; it returns a choice — which button to press, which bot to route the task to, whether a human needs to respond.
🔹 The most useful cases aren’t games but boring routine: classifying comments, finding a file by its description, checking code before review.
🔹 The author honestly warns: “fast” ≠ “safe,” and a low price per call ≠ cheap at scale.
Useful for anyone thinking about how to integrate AI into workflows rather than just chatting with it.
📖 [https://t.co/tyEA6QQPlD]
#AI #Jev #TypeSafe #automation

[にく @29meat_ai]

https://t.co/c1uZHe8zIu

[developer's journal @DevJournaI]

XGEN Labs announced a technical prototype of Generative World Simulation. The system connects the JING model (first-person experience) with the DAO engine (shared world state and rules). XGEN-JING ranked #1 on WBench. GitHub and Hugging Face links in the original tweet. 👇

[XGEN Labs @XGEN_labs]

🔥 Today, we are truly excited to announce our technical prototype, the Generative World Simulation system, which integrates JING（镜）, an interactive experience model, with DAO（道）, a computable shared-world engine.

The coupled model and engine connect first-person experience with a shared world that continues to evolve beyond any individual observer.

Conditioned on actions and observation history, JING enables agent navigate, manipulate, and communicate from a first-person perspective in the world. Watch our demo video to see it in action!

On the official WBench leaderboard as of September 17, 2026, XGEN-JING ranked #1 on the Full split, and #2 on the Navi split. 🎉

DAO maintains shared world state and rules, computes the consequences of actions, and provides JING with only what the current observer can perceive. It also supports autonomous agent decision-making, enabling agents to act independently within an evolving shared world.

Together, DAO and JING move beyond generating the next frame toward simulating the world behind it. This marks a small step towards OASIS: not just a world that responds to you, but a world—and a society—that evolves with and without you. 💪

Explore XGEN Labs～:

🔗 Website: https://t.co/tqTvPIk19B

�� HF: https://t.co/wxLRmJNYgO

🦊 GitHub: https://t.co/2DMvURLUsK

[developer's journal @DevJournaI]

NVIDIA Research presents Axolotl3D: a unified, occlusion-aware framework for faithful 3D shape completion from images, camera data, and partial geometry. SOTA across single- and multi-view settings. Accepted at #ECCV2026.
📄 https://t.co/6P60E2wmnB
🌐 https://t.co/XsU2VC6jYg

[NVIDIA AI @NVIDIAAI]

Images rarely show an object’s full 3D geometry.

At #ECCV2026, our research team introduced Axolotl3D, a multimodal and occlusion-aware 3D generation model. It combines images, camera data and partial geometry to reconstruct missing regions while preserving observed ones, achieving state-of-the-art results across single- and multi-view settings.

Project page: https://t.co/bC156qkpl1

[NVIDIA AI @NVIDIAAI]

Images rarely show an object’s full 3D geometry.

At #ECCV2026, our research team introduced Axolotl3D, a multimodal and occlusion-aware 3D generation model. It combines images, camera data and partial geometry to reconstruct missing regions while preserving observed ones, achieving state-of-the-art results across single- and multi-view settings.

Project page: https://t.co/bC156qkpl1

[developer's journal @DevJournaI]

Worldsplat on Product Hunt: sketch in 3D, generate the scene instantly.

Looks like a fun way to visualize concepts without the usual 3D pain.

Check it out 👇
https://t.co/0UoFw3c0XA

#3D #AI #ProductHunt

[developer's journal @DevJournaI]

No more scrolling through Giphy. Describe the exact reaction — get it in moments.👀

[developer's journal @DevJournaI]

What if every chat had an AI GIF button?
Type the reaction you want, generate it, and attach it without leaving the conversation.
I created this using MiniMax H3 Max via @fal. ⚡️
Would you use this? 👀

[developer's journal @DevJournaI]

@SouthernShotty

[developer's journal @DevJournaI]

@mark_lundin @playcanvas

[developer's journal @DevJournaI]

@AudienceCapital hi

[developer's journal @DevJournaI]

Guys. This is really cool—you have to check it out and give it a try.

[Christopher J. DiMarco @chrisjdimarco]

My next Astra experiment: create a mini diorama motocross simulator.

Set riders skill levels and sculpt the course.

Used Astra Medium/High depending on tasks. Burned through 70% of my weekly limit already 😆.

Play Mini Motto for free: https://t.co/Th5Ni3J906

[developer's journal @DevJournaI]

@chrisjdimarco That's really cool. I'm thrilled👏👏👏👏

[developer's journal @DevJournaI]

Great news for those who animate in Blender! 🆓 @ehsanAnim released a convenient set of sliders for Blender. A set of 16 free tools for quick setup of timing and interpolation. 👇

[Ehsan @ehsanAnim]

⭐️ Free Tool for Blender Animators ⭐️

It's been a while since I've been thinking about making something like this for Blender, so I could have an easier time animating in this software.

It's a set of 16 sliders from my Maya Animo toolset.

[developer's journal @DevJournaI]

@tugg_speeedman But we'll keep an eye on it and post the code as soon as it becomes available.

[developer's journal @DevJournaI]

This is exactly what ViSculpt solves! 🤯

Instead of clumsy Python scripts, it uses AI agents (VLMs + SAM2) to operate Blender's GUI directly. It touches the "bare metal" to physically sculpt on existing 3D meshes just like a human artist.

📄 arXiv:2608.24169 #Blender #3D

[Shikhar @xikhar]

@alexzenciks "Using coding agents to control blender has that communication overhead because it is a complex software and was meant for humans to use.

We need an agent to touch the bare metal of 3d surface in blender.*

[developer's journal @DevJournaI]

@tugg_speeedman Unfortunately, it's not available yet. The article says the code will be released later. 😟

[developer's journal @DevJournaI]

We like it, and we support this idea😻

[Yun @Yun_HDY]

you can make a 15s motion with only one prompts

cozyclay nextupdate will be release at 9/10

https://t.co/CxgZHabgiI

[developer's journal @DevJournaI]

https://t.co/HaqKiAgHY6