Metarix
AI & Software Intelligence for Builders
Research, benchmarks, and technical analysis for people building, evaluating, and investing in modern software.
Featured Intelligence
AI Agents · News · Alex Rivera · Sep 18, 2026 · 14 min
Factory Is Now Worth $5 Billion
The enterprise coding-agent company raised $200M and tripled its price to $5B. The product pitch is not one more chat. It is a software factory you can put behind a firewall.
Read article →
Benchmarks · Analysis · Alex Rivera · Sep 18, 2026 · 20 min
Claude Now Does a Quarter of Anthropic's Work
As of August, Claude leads 26% of Anthropic's AI research work. Humans still watch. Nothing runs fully on its own. The useful part is the dashboard, not the headline.
Read analysis →
AI Agents · Analysis · Alex Rivera · Sep 17, 2026 · 17 min
Cursor Runs the Fleet Now
Projects is a coordinator that does not write the code. It keeps context for months, farms subagents in the cloud, and watches Slack and CI without a new prompt.
Read analysis →Latest Intelligence
View all intelligence →SaaS · News · Alex Rivera · Oct 23, 2026 · 8 min
Claude Enterprise Merges Chat and Cowork
Claude Enterprise now runs chat and Cowork as one experience, with agentic work in Anthropic-managed cloud VMs. Admins get a new Efficient mode, per-role model entitlements, spend limits and pooled budgets to keep the token bill in check.
Read article →AI Agents · News · Alex Rivera · Oct 23, 2026 · 8 min
Codex CLI 0.164 Starts Masking Your Credentials
OpenAI's Codex CLI 0.164.0 masks credentials in supported remote Bash and Zsh snapshot sessions and adds sign-in for enterprise MCP servers. It also brings ten prompt presets, inline /model and /permissions arguments, and fixes for compaction and session resume.
Read article →
Productivity · News · Alex Rivera · Oct 23, 2026 · 8 min
GitHub Apps Can Now Replace Enterprise PATs
Enterprise installations for GitHub Apps are generally available, with new permissions for policies, rulesets, 2FA, verified domains, IP allowlists and access restrictions. Most of the reasons to keep a classic PAT for enterprise admin work just went away.
Read article →
Cloud · Analysis · Alex Rivera · Oct 23, 2026 · 8 min
Google Cloud Now Lets Anthropic Run Claude
Claude on Google Cloud now has an Anthropic-managed option that gets Managed Agents, skills and code execution on day zero. The trade is that Anthropic, not Google, becomes your data processor.
Read analysis →
AI Agents · Analysis · Alex Rivera · Oct 7, 2026 · 13 min
Atlassian Wants Agents in the Same Loop
AMP gives agents identity, scoped authority, and the Teamwork Graph. Jira now takes Devin, Factory, and Warp. Codex is next. Ten million human-agent collabs a month, they say.
Read analysis →
Cloud · News · Alex Rivera · Oct 6, 2026 · 10 min
Vercel Put Le Chonk in the Gateway the Same Day
mistral/mistral-large-4 on AI Gateway, no Mistral account required. $0.68 in, $2.09 out. npx vercel ai-gateway setup wires Claude Code.
Read article →Topics
Five intelligence buckets
AI Agents & Coding
Autonomous coding systems, Cursor workflows, LLM developer APIs
28 articles
Latest: Claude Enterprise Merges Chat and Cowork
Explore AI Agents →Cloud & Architecture
Serverless, edge computing, CMS setups like Sanity, deployment pipelines
6 articles
Latest: Google Cloud Now Lets Anthropic Run Claude
Explore Cloud →Developer & Productivity Tools
Command-line utilities, text editors, efficiency scripts
17 articles
Latest: Codex CLI 0.164 Starts Masking Your Credentials
Explore Productivity →SaaS Ecosystem & Business
Software monetization, indie hacking tech stacks, market analysis
15 articles
Latest: Claude Enterprise Merges Chat and Cowork
Explore SaaS →Benchmarks & Deep Dives
Rigorous performance testing, framework comparisons, data-backed architectural breakdowns
7 articles
Latest: Kimi K3.1 Pro Closes the Open-Weight Gap
Explore Benchmarks →Research
Explore research →
AI Agents · Benchmark · Alex Rivera · Oct 23, 2026 · 10 min
Kimi K3.1 Pro Closes the Open-Weight Gap
Moonshot AI's Kimi K3.1 Pro ships with a 1M-token context, $2/$10 per million tokens, and open weights promised by November 4. Independent testing puts it fourth overall and first on Terminal-Bench 4.0.
Explore benchmark →
Benchmarks · Benchmark · Alex Rivera · Oct 6, 2026 · 14 min
Mistral Named the Trillion-Param Model Le Chonk
1 trillion parameters, 49 billion active, trained on 3,800 GB200s in Europe. Preview API today. Weights by the end of October. Cyber is the pitch.
Explore benchmark →
Benchmarks · Benchmark · Alex Rivera · Sep 17, 2026 · 18 min
Agents Get a Trustmark
AIUC-1 is trying to be SOC 2 for agents: thousands of evals, a certificate, and now a $40M Series A to sell audits and insurance. A badge is not a kill switch.
Explore benchmark →Essential Reading
AI Agents · Comparison · Alex Rivera · Sep 17, 2026 · 14 min
Salesforce Splits the Stack: Koa In-House, Claude at the Door
Koa is a CRM reasoner Salesforce trains and hosts. Claudeforce puts 37 sales skills inside Claude. Same Dreamforce week; two control models.
View comparison →AI Agents · Deep dive · Alex Rivera · Sep 17, 2026 · 16 min
A Map of AI That Improves Itself Is Not a Launch Date
A ByteDance–Tsinghua–Shanghai AI Lab preprint ranks 491 papers on five rungs of recursive self-improvement. The interesting part is what the ladder does not measure.
Read the deep dive →SaaS · Analysis · Alex Rivera · Sep 16, 2026 · 38 min
Siri AI Is Finally Shipping. Treat It as OS Infrastructure, Not Another Chatbot.
Apple’s rebuilt assistant is in beta with iOS 27. The story is not a smarter Siri voice — it is an on-device plus Private Cloud Compute stack, a hardware and geography gate, and a system API surface developers have to take seriously.
Read analysis →AI Agents · Analysis · Alex Rivera · 20 min
The Death of Chat: Why Type-Safe, Machine-Native AI Is Changing Software Engineering
Co-inventors of ChatGPT’s foundational RLHF have launched TypeSafe AI, shattering traditional LLM latency and cost limits with 'Jev'—a type-safe, machine-native model built to make instant, structured decisions directly inside codebases.
Read analysis →Most Read
- 01Claude Drops Extra Tabs
AI Agents · News · Alex Rivera · Sep 16, 2026 · 20 min
- 02Claude Code Now Runs a Team
AI Agents · Analysis · Alex Rivera · Sep 19, 2026 · 15 min
- 03Zed Turned Off Pull Requests
Productivity · Analysis · Alex Rivera · Sep 19, 2026 · 14 min
- 04The Agent Now Runs in Your Container
Productivity · News · Alex Rivera · Sep 19, 2026 · 13 min
- 05Hang Ten Sells Outcomes
SaaS · News · Alex Rivera · Sep 17, 2026 · 20 min
Get the Signal
Short analysis of AI tools, architecture, and software markets. We store your email for when sending is live.
Saved for when we start sending. Privacy Policy