Welcome back. Remember that mystery model crushing benchmarks last week? It finally has a name and an official launch. Read on for who built it and what it could mean for frontier models. Meanwhile, Meta’s big bet on agentic AI is reportedly backfiring, with Reuters saying its plans for massive cuts have run into trouble.
Also: Meta engineer shares how the promotion ladder works, a command to visualize and learn anything, and a new model that can spin up real-life videos with one prompt.
Today’s Brief
Core skills you need to be an AI engineer
20 essential GrokBot hacks from devs
A laid-off AI engineer exposes major industry flaws
How to build adaptive agents (tutorial)
TODAY IN PROGRAMMING
Click here to see ox-alpha's full benchmarks.
The anonymous model devs loved just got a name: For the past week, coders have been piling onto "Ox-alpha," a surprise entry at the top of the OpenRouter and OpenCode charts. Z.ai, the team behind GLM, just revealed it as GLM-5.3-Flash, an open multimodal system built for agentic workflows and long-horizon software engineering. The lab claims it beats GLM-5.2 across its benchmarks at a tenth of the price and lands within half a point of Claude Opus 4.8 on the lab's own private coding benchmark.
Meta's 'AI native' dream hit its own messy code: Zuckerberg kicked off Project OT this year, shrinking teams into tiny AI-equipped pods while eyeing up to 60% cuts, according to Reuters. It didn’t go to plan. While code changes surged 220% YoY, actual features reaching users only grew by 36%. Unchecked agents caused constant disruptions, incidents spiked by 40%, and firefighting time climbed 70%. We even saw the chaos in June when hackers exploited Meta's new AI support bot to hijack high-profile Instagram accounts. Meta has since scrapped the November layoffs after the team pushed back, but the damage is done. Morale is tanking, and talent is walking out the door.
Google's new model turns rough speech into clean transcripts: Google just dropped Gemini 3.5 Transcribe, now in public preview for devs via the Gemini API and Antigravity. It gives engineering teams a model built for voice agents, live captioning, and post-call analytics. It strips out filler words, adapts to custom jargon, tags up to three speakers, and hits a 2.6% word error rate on non-streaming audio, according to Artificial Analysis. Try it here.
PRESENTED BY MONGODB
A new Managed MCP Server connects Claude Code, Codex, Grok Build, and Devin to MongoDB Atlas, giving coding agents direct access to live operational data.
INSIGHT
What does an “AI engineer” actually need to know?
Source: The Code, Superhuman
The title outran the definition. The "AI engineer" title is popping up on job boards everywhere, but there’s still zero consensus among teams on what these engineers actually do. It leaves most of us guessing which skills are worth leveling up and which are just hype. To clear the air, ML professor and researcher Andrew Ng dropped his AI Engineering Skills Map. Built from over 10,000 job postings and dozens of structured interviews with hiring managers and recruiters. He broke down the core competency of building and deploying AI apps into six distinct pillars:
LLM foundations. This explains how a model reads a prompt and produces an answer. Once you grasp that concept, it becomes clear when you should trust the model, where it will fail, and when a custom build is actually worth the effort.
Grounding models with data. A model is only as good as its context. There are plenty of ways to supply that context, and mastering which one to use is the real secret.
Building agentic systems. Some agents follow fixed steps while others decide their own next move. The real skill is choosing which tools they can use and setting limits that keep them in bounds.
Evaluation-driven development. You need to prove that a change actually made things better instead of just trusting a gut read. This means you should test outputs with code, another model, or a human doing the checking.
Operating in production. Operating in production involves keeping a live system healthy. This means watching how it performs, spotting when quality slips, blocking bad inputs, and managing both cost and speed.
Machine learning foundations. This is the math that LLMs are built on. You need to understand the base layer instead of just focusing on the model sitting on top of it.
Where to start. This list might feel like a lot to take in, but it’s actually pretty straightforward. The devs who really stand out are the ones who get the foundations down first and then build a solid evaluation loop.
P.S. If you’re ready to start, DeepLearning.AI's course catalog maps onto all six skills.
PRESENTED BY BUILDKITE
Devs finally have a challenger to GitHub code hosting with Origin.
We also run CI for OpenAI, Anthropic, Mistral, xAI, Airbnb, Nvidia, Canva, Uber, Meta, Pierre, Bun and more.
Build faster, more securely, more reliably, with the CI engineered for frontier scale.
Start with our 30-day all-access trial.
IN THE KNOW
What’s trending on socials and headlines
Meme of the day.
Grok Bot Hacks: A CTO’s Grok Bot already runs his inbox, meeting notes, and phone calls in Portuguese. He shared every hack he uses, plus one warning most users miss (2.9K bookmarks).
Inside Look: A viral post from a laid-off AI engineer exposes major industry flaws he witnessed firsthand while working on training models (1.8K likes).
Promotion Ladder: An ex-Meta staff engineer grinded as a junior but never got promoted. He now says leveling up comes down to just one sentence, and it's not about hard work.
Speed Demo: Devs can now spin up high-quality video clips on this model with a single prompt. One senior dev tested it out and called the results "insane" (1.3K likes).
Visualize Outputs: OpenAI's dev team just demoed a one-word command in ChatGPT Work and Codex. Trigger it to transform those intimidating walls of text into something easily digestible (2.2K likes).
Nuked Machine: A blockchain CTO asked Claude to build an agent sandbox, a routine task for most devs. It wiped his entire home directory. Check out what went wrong (4.3K likes).
TOP & TRENDING RESOURCES
Click here to watch the tutorial.
Top Tutorial:
How to build adaptive agents (by Oracle): You'll learn three ways to stop agents from relearning the job every session. This tutorial covers turning an agent's own traces into reusable, human-approved skills, building a code knowledge graph for smarter retrieval in large codebases, and knowing when it's worth adapting the model itself.
Top Repo
OpenWorker (16.4K ⭐️): An open-source agent that completes tasks on your laptop, now with built-in cybersecurity agents for vulnerability, dependency, and cloud config scanning.
Trending Cookbook
How to build agent environments & tasks (by LangChain): A guide to building synthetic agent environments. You'll learn a two-step pipeline: first, how to turn traces and code into a spec, then how to convert that spec into a full eval task. They've packaged everything into an updated eval-engineering skill so any team can run it.
AI CODING HACK
How to run git commands inside Codex without burning tokens
Asking Codex to run git status costs tokens and a wait while it decides to run the command and narrates the result. A dev replied to a viral joke tweet with the fix: Codex has a built-in shell prefix.
Step1: Prefix any line in the Codex prompt with
!and it runs locally, right away.
!git statusStep 2: The output lands in the conversation, so your next prompt can use it without pasting anything.
Summarize these changes and flag anything I shouldn't commit.Commands respect your sandbox and approval settings. Full shortcut list in the Codex CLI reference.
P.S. Get 50+ AI coding hacks for Claude Code, Cursor, and Codex here.
IN CASE YOU MISSED IT
Our most-clicked story from yesterday
Shopify's CEO is threatening to ban Claude Code because it won't read AGENTS.md. Anthropic responded, and the whole exchange is going viral.
Grow customers & revenue: Join companies like Google, IBM, and Datadog. Showcase your product to our 300K+ engineers and 150K+ followers on socials. Get in touch.
What did you think of today's newsletter?
You can also reply directly to this email if you have suggestions, feedback, or questions.
Until next time — The Code team
