Nemotron 3 Nano Omni was designed for powering subagents. Instead of stitching together separate models for language, vision, and speech, it ties them into a single architecture that more efficiently feeds context to orchestrators.
AGENTS
-
Building DeepAgents Examples: Community Feature Requests
By
–
I’m going to build a bunch of deepagents examples (using deepagent deploy) over the next few days What examples would people want to see?
-
New Method to Train and Evaluate AI Agents Efficiently
By
–
Vibe train your AI agents. There's a new method that could replace LLM-as-a-judge for production agents. Most teams rely on a giant LLM as a judge to evaluate and guard their agent. But it has two major drawbacks: – It's slow and expensive at inference time
– It often misses -
The Rise of AI Native Companies and Automated Workflows
By
–
The rise of AI Native Companies.
@IrenaCronin and I write this newsletter every week.
AI native companies are businesses designed around AI from the beginning rather than adding AI to old workflows. They use smaller teams, automated processes, and AI agents to move faster, -
Base44 launches agent platform with migration and UI tools
By
–
OpenAI and Anthropic are racing to build the smartest agents.
— Chubby♨️ (@kimmonismus) 28 avril 2026
Base44 is building the platform those agents actually need to run on. While the labs ship models, Base44 just shipped: One-click migration from 6 major platforms Schema reconstruction Custom UI generation Agent… https://t.co/76DZL5x3XdOpenAI and Anthropic are racing to build the smartest agents. Base44 is building the platform those agents actually need to run on. While the labs ship models, Base44 just shipped: One-click migration from 6 major platforms Schema reconstruction Custom UI generation Agent
-
Google Kaggle Launch Free AI Agents Vibe Coding Course
By
–
This is MASSIVE for vibe-coders. @Google and @Kaggle are officially bringing back their 5-day intensive vibe-coding course with AI agents And yes, it’s completely FREE. Last time, they had 1.5 million learners, which means something! Here’s what you’ll be learning: →
-
Prompt drift in parallel terminals fixed by hand-built markdown system
By
–
Translation: prompt quality without persistent context produces drift. A solo operator running parallel Claude Code terminals put it best: "Each terminal has no idea what decisions I made in the other." Their workaround? A markdown file system they built by hand.
-
Structured decision files boost agent performance over raw logs
By
–
The 7-agent experiment confirmed it directly. Agents that wrote structured DECISIONS.md and PROGRESS.md files outperformed agents that dumped raw logs. Same models. Same prompts. Same starting budget. The variable was the thinking architecture between sessions.
-
Indie builder threads on AI memory and agents converge
By
–
In the last 7 days, two threads from indie builders hit 100+ comments each. One asking how to handle memory across AI tools. One running 7 AI agents head-to-head with $100 each. Both threads converged on the same finding.
-
Builders admit LLMs can’t think, manually write thinking architecture files
By
–
Hundreds of builders just admitted LLMs don't think. Their fix: manually writing thinking architecture files (PROGRESS.md, DECISIONS.md, IDENTITY.md) for every project. Most AI advice still chases better prompts. The real work moved up a layer. Here's what changed and what it