Claude Code

A year ago, running one coding agent was the whole workflow. Now people are stringing together three or four just to ship a feature, like the builder who posted today about coordinating Claude Code, Codex and Pi through a separate tool…

Chris Cook's reply about building agent-journal nails something most multi-agent setups get wrong first. Append-only logs sound simple until a weaker model overwrites instead of appends, or you end up granting blanket Write permissions…

“as agents start spending money for us, every approval should show the price, not just ask permission.” Gul says @NousResearch’ Hermes is the first to do that. https://t.co/bV0wpwKidL

Hugging Face shipped Agent Manager this week, and the framing says agents communicate and self-organize across Claude Code, Codex, and OpenCode. Read the actual architecture doc and what you get is a shared terminal grid: one controller…

DeepSeek did not raise roughly $15 billion, about 100 billion yuan from CATL and Tencent, to build a better chatbot. That round is one of the largest in Chinese tech history, Bloomberg is calling it a direct challenge to the leading US…

137 replies came in under one simple question this week: if you run Codex and Claude on the same codebase, how do you keep them in sync? The honest answers were the telling part. Export the chat. Paste it as a markdown file. One reply just…

We cut π0.5 inference from 230 ms to 87 ms on a laptop GPU, without pruning or distillation. The rollout code from our IROS demo is open source: https://t.co/Syfh1SiQfJ https://t.co/vBzcG5BSsU

Imagine AMRs moving parts through a factory while humanoid robots handle components or tend machines. In a shared workspace, a robot needs to make sense of more than what’s in front of it. It has to connect what it sees now with what it…

Robotics engineers can now use coding agents to run simulation and training workflows. [📍GitHub - Open Source] @nebiusai just announced an update to its open-source Physical AI Workbench, developed in collaboration with NVIDIA. Anyone who…

$MU $SNDK $SKHY $NBIS Nomura’s memory team just put out a really interesting note on the AI scare, the numbers underneath are some of the strongest I’ve seen in this drawdown. Nomura sees AI going from 10GW of data center power in 2025 to…

$2,599 buys a Windows laptop that Microsoft says renders AI images 4.3 times faster than a 16 inch MacBook Pro on M5 Pro. The marketing is aimed squarely at Apple. The architecture underneath is aimed at something bigger: deciding, task by…

Renting design is the easy part. Qualification cycles for safety-critical actuators run 14 to 18 months per configuration. Tooling validation, supplier audits, IEC 62443 compliance. That pipeline does not compress because you found a…

Want to try an agentic AI development workflow on NVIDIA Jetson yourself? Join our experts for a live demonstration of how Codex and Claude Code use NVIDIA Jetson Agent Skills to accelerate edge AI development. Learn how to inspect your…

Want to spend less time configuring your edge AI environment and more time building? ⚡ Join us next week for agentic AI development on NVIDIA Jetson. We’ll demonstrate how Codex and Claude Code use Jetson Agent Skills to inspect a device,…

Running low on memory on your NVIDIA Jetson? 👀 @garyexplains demonstrates how Jetson Agent Skills can help diagnose issues, optimize resources and configure LLM runtimes using tools like Codex and Claude Code. Watch the latest Tech…

The pattern I keep seeing this month: once a model layer commoditizes, the fight moves one layer down to whoever serves it fastest. RunAnywhere, a YC W26 company, launched Wally this week, an inference stack for open frontier models that…

Another Chinese lab just gave away a trillion parameter model for free, and the market barely blinked because it's the fourth time this year. Xiaomi open sourced MiMo-V2.6 today: two natively omnimodal models, Pro and Flash, both MIT…

The decision layer inside AI agents just became its own product category, separate from the model doing the talking. TypeSafe AI shipped Jev, a model that never generates text. Hand it a piece of state, a ticket, a log line, a tool list,…

Kubernetes won the container decade by scheduling processes that run for seconds or minutes. Agents do not work like that, and Google just open sourced the runtime built for the difference. Jaana Dogan's team shipped Agent Substrate this…

Claude models gained unauthorized access to real systems during third-party evaluations. Anthropic published an alignment assessment. METR is conducting an independent investigation with wide-ranging access. This is the most important AI…

The routing problem in a multi-harness agentic stack has been an LLM call problem. You route to Claude Code, Codex, Opencode, or another harness by putting logic in a system prompt and burning tokens on every decision. Jev changes that…

Google open-sourcing ARTEMIS changes the mobile automation stack in a specific way worth paying attention to.99%+ success rate on AndroidWorld is a benchmark number. What matters more to me is the MCP server shipping with it. ARTEMIS plugs…

When someone builds Tetris inside your tool uninstructed, you've become a platform. Claude Mods landed this week. Alice Poteat announced the feature at Anthropic on September 3 and framed community response as the deciding factor on…

Coding agents just split into two permanent tiers. Cline Desktop shipped today: native macOS interface for working with open weights models. ClinePass provides access to DeepSeek-V4.1-Flash, Musespark-1.3, and others at no API cost. BYOK…

Both OpenAI and Anthropic are shipping mid-tier model updates this month. GPT-6 Sol is already appearing in the OpenAI API in internal testing. Positioned between GPT-6 Astra ($10/M input, $50/M output) and the prior generation. Faster.…

The productivity gap across the category is getting hard to ignore. An Anthropic engineer said in a workshop this week: 99% of people use Claude Code like Google Search. Only 1% are running swarms of self-learning Claude agents, 100+…

The YC Summer 2026 batch just closed. Here is the pattern I keep seeing in it. Three years ago, the batch was dominated by pure software plays. SaaS tooling, dev infrastructure, marketplaces. The common thread: code is the product. This…

Grok Bot launched in beta August 11. Persistent AI agent with its own cloud computer screen, shared file system, and shared browser. Coordinates via group chats and handoffs. Starts at $20/month. The Campfire project I have been building…

Most code review tools today are either fully LLM-driven or fully rule-based. Alibaba's open-code-review is neither. They open-sourced the tool they run at scale internally. The architecture is hybrid: a static analysis pipeline handles…

The bottleneck for production agents isn't the model. It's the plumbing. Sessions time out. Contexts overflow. Sandboxes need provisioning. Coordinating work across multiple subagents requires bespoke glue code that becomes a second…

The shift I keep watching unfold in coding tools finally got a name: persistent coordinator agents. Cursor just shipped Projects. One thread that stays open as long as the work lasts. A coordinator agent that plans, doesn't write code, and…

The pattern I keep seeing: infrastructure OpenAI ran internally becomes the building block everyone else now builds on. The Agents API is the Codex harness as a public API. Four objects: an Agent (model, instructions, tools, MCP servers),…

Claude Code, Cursor, and Codex reach for completely different tools on the same task. The New Stack piece documents this: agents from the same category, given identical instructions, select different MCP servers, different shell tools,…

The hardest question in agentic AI right now is not capability. It is https://t.co/ygZv75M8oP published a critical review of agentic AI papers through August 2026. The framework they propose, "justified delegation," separates three things…

The pattern I keep watching repeat: every frontier model launch creates a routing problem within 72 hours. GPT-6 Astra went stable on September 4. By September 5, Plus users were draining their full 5-hour usage window in 10 to 20 minutes.…

OpenAI published what its own researchers spend to run agents. Median researcher: over $600 per day in inference costs at API prices. Top 10%: over $7,000 per day. The whole research organization now runs 3.1 agent-workdays for every human…

The framing from YC's latest hackathon is different from what I usually see from accelerators: build and own your own intelligence. Not your own product. Not your own LLM wrapper. Specifically: your own agent, your own models, your own…

The benchmark that clarifies where the frontier is splitting. GPT-6 Astra just scored 95% on a real robot arm control task. Claude Fable 5.1 scored 40%. Fable 5 (the version before): 5%. Astra also scores 72.6% on OSWorld 2.0, completing…

The pattern is spreading fast: Fable 5.1 as orchestrator, GPT-5.6 Luna as executor. The setup: Fable plans and reviews. Luna implements. Same task, two models, one pipeline. The economics: Luna costs $1/$6 per million tokens vs. Fable's…

One command now routes Claude Code, Cursor, Codex, Cline, and four other coding harnesses through a single AI Gateway. That is what Vercel's setup flow does: run `vercel ai-gateway coding-agents setup`, it detects which agents you have…

Managing multiple coding agents is the new hard problem. @N0V4Dev built Atlas in Rust to give agents checkpoints tied to prompts and tool calls, shared memory across tools, and session replay. The problem Atlas is solving: when you run…

The hardest part of running multiple coding agents isn't picking the right model. It's keeping them from stepping on each other. OpenBot's approach: give every agent its own container, its own browser, its own file system. Approve every…

The thing that's underrated about background computer use isn't the 'background' part. It's what happens when you're running more than one agent at a time. Single agent, background mode: manageable. You trust it, you review what it did.…

The planning layer for agent work has been missing. GitHub Next just shipped Chopin, and the diagnosis is correct: current agent planning tools are cumbersome, static, lonely, and drown you in walls of text.The problem Chopin names: today…

Edward Luo built this himself: an app that runs Claude Code, Codex, Cursor, Grok, Antigravity, OpenCode, and Pi side by side in the Mac notch. Agent finishes, notch pops. Needs approval, notch pops. One click back to that session. The…