ContentBuffer
Meta's AI app Muse has hit 10 million users, with 25% of them using the app daily. The app's agentic shopping feature, powered by Plaid's connector, sees users spending an average of over $200 per purchase. This makes Muse a significant player in the AI-driven commerce space. Muse's security features, including Muse Secure VM and the upcoming Muse Confidential VM, ensure user data privacy. Meta plans to monetize Muse through a freemium model, with advanced AI models and more tokens available for a fee. The company forecasts Muse to generate more revenue per user than its 'family of apps' users, currently around several hundreds of dollars annually.
A year-old startup has just raised a $1 billion Series A, pushing its valuation to $6 billion. This is a massive boost for a company that launched with a $180 million seed round just nine months ago. The company's founders, who include a former Thiel fellow, a Neuralink alum, and a pair of Stanford PhDs, are building what they call a 'megakernel framework' designed to train and run intelligent systems more efficiently. This framework can perform entire training runs, serve inference, and even play video games simultaneously, all while overlapping memory-bound inference with compute-bound training. The $1 billion influx will be used to hire more talent and accelerate their ambitious goal of creating models that are a million times more data efficient. Let's dive deeper…
In today's ContentBuffer update:
Tesla's Optimus Robot Production Begins, Q3 Profits Drop 43%
Recursive's Flint: A Free Coding Model for Laptops
Automattic Layoffs: CEO's Private Slack Post Sparks Controversy
GM's 4th-Gen EV Platform Centralizes Computing
5 new AI tools & 5 new AI jobs
More tech news
Latest Development
Image source: techcrunch.com
Summary: Meta's Muse AI app has hit 10M users, with daily active users driving $200+ average spend. The app's agentic shopping feature and robust security measures make it a key player in AI-driven commerce.
Key Points:
Muse has over 10 million users, with 25% using the app daily.
Over a quarter of Muse's engaged users spend over $200 per purchase.
Muse uses Plaid's connector for agentic shopping, a first for the company.
Muse Secure VM protects user data, with Muse Confidential VM to block even Meta's access.
Meta forecasts Muse to generate more revenue per user than its 'family of apps'.
Why it matters: If you're a developer working on AI-driven commerce solutions, Muse's adoption and average spend figures are a clear indicator of the market's appetite for such tools. The app's integration with Plaid's connector and its security features like Muse Secure VM and the upcoming Muse Confidential VM highlight the importance of robust data protection in AI applications. This is especially relevant for developers looking to build similar platforms.
Image source: techcrunch.com
Summary: A year-old startup raised a $1 billion Series A, quadrupling its valuation to $6 billion. The founders are building an AGI lab with a focus on data efficiency.
Key Points:
Startup raised a $1 billion Series A, pushing valuation to $6 billion.
Round was co-led by Index and Lux, with earlier backers Sequoia, GV, and Kleiner Perkins.
Company was founded by three individuals: a former Thiel fellow, a Neuralink alum, and two Stanford PhDs.
Megakernel framework can perform training runs, serve inference, and play video games simultaneously.
Goal is to create intelligent systems that cost less to train and run, aiming for a million times more data efficiency.
Why it matters: If you're a researcher or engineer working on AGI or data-efficient models, this startup's $1 billion round and megakernel framework could change the game. The framework's ability to overlap memory-bound inference with compute-bound training could significantly reduce costs and improve efficiency. However, the $1 billion investment is a double-edged sword: while it accelerates development, it also raises the bar for competitors.
Image source: techcrunch.com
Summary: Tesla has started producing its Optimus robot while Q3 profits dropped 43%. The financial impact of such ambitious projects is clear.
Key Points:
Tesla's Q3 revenue was $28.04 billion, down slightly from $28.23 billion in Q2.
Tesla delivered 486,532 vehicles in Q3, a slight increase from Q2 but down from Q3 2025.
Tesla's Q3 profits fell to $787 million, a 43% drop from the same period in 2025.
Tesla's services business grew 33% year-over-year, contributing $4.6 billion in revenue.
Tesla's Optimus robot production began at the Fremont factory, marking a significant milestone.
Why it matters: Tesla's Optimus robot production is a significant step towards commercial viability, but the company's Q3 financials highlight the financial risks. If you're tracking Tesla's robotics efforts, the Optimus robot's production is a clear signal of progress. However, the company's Q3 profit drop to $787 million underscores the financial challenges of such ambitious projects.
Image source: techcrunch.com
Summary: Recursive unveiled Flint, a free coding model for laptops, built with the help of an AI-driven research system. It's a big deal for developers looking to experiment with AI coding without hefty compute costs.
Key Points:
Flint, released by Recursive, is a free coding model that runs on laptops.
The Eureka Machine, an AI-driven research system, contributed to Flint's development from pre-training to inference.
Flint uses only 2.5 billion of its 40 billion parameters at a time, making it smaller than Meta's 816 billion-parameter Muse Spark Open.
Flint outperformed models under 50 billion parameters on SWE-Bench Verified and scored higher than most models on Recursive's own tests.
Recursive's compute deal with AWS in June was less about headcount and more about agent count, highlighting the importance of AI in research.
Why it matters: If you're a developer looking to experiment with AI coding assistants, Flint is a game-changer. It's free, runs on a laptop, and outperforms many larger models. However, it requires more guidance than paid assistants, making it ideal for those who want to dip their toes in without a steep learning curve.
Image source: techcrunch.com
Summary: Automattic begins layoffs, impacting 1,397 employees, and the CEO's private Slack post about a departing exec's health condition sparks controversy.
Key Points:
Automattic, parent company of WordPress.com, WooCommerce, and others, initiates layoffs affecting 1,397 employees.
The company's interim CFO left earlier this month, and the CEO is seeking external candidates to fill the role.
The layoffs impact various teams, including communications, social, and VIP enterprise business.
Automattic plans to spread out the layoffs over the coming months to improve Q4 revenue figures.
The company's focus on PHP security issues and a debt renegotiation in June add to the uncertainty.
Why it matters: Automattic's layoffs and internal communication mishap affect teams across the company, including communications and VIP enterprise business. The CEO's private Slack post about a departing exec's health condition has raised concerns among employees about transparency and company culture. The layoffs and debt renegotiation in June signal financial challenges for the company.
Image source: techcrunch.com
Summary: GM unveiled its 4th-gen EV platform, focusing on central computing and novel battery tech. The Cadillac Escalade IQ will debut in 2028 with a 40x more powerful CCU, handling propulsion, chassis, perception, and safety. A big deal for automotive software.
Key Points:
GM's 4th-gen EV platform centralizes vehicle functions on a CCU 40x more powerful than current systems.
The platform will debut in the Cadillac Escalade IQ in 2028, rolling out autonomous highway driving the same year.
GM's capitalized software amortization expense in 2025 was $866 million, highlighting the company's software focus.
The new platform will centralize much of the vehicle's functions, trimming wiring by 30% and simplifying assembly.
GM's revamped development strategy includes using enterprise AI tools to speed up vehicle development and save costs.
Why it matters: If you're working on automotive software, GM's 4th-gen EV platform is a game-changer. The central computing unit (CCU) is 40x more powerful than current systems, handling propulsion, chassis, perception, and safety. This centralization simplifies wiring by 30%, trims weight, and paves the way for more advanced autonomy features. The Cadillac Escalade IQ will debut in 2028 with this tech, and GM plans to roll out autonomous highway driving the same year.
New Tools & Job
MachineTranslation.com - MachineTranslation.com is an AI-powered translation platform that translates text and documents by comparing outputs from multiple leading AI translation models. Its consensus technology helps users choose the most accurate translation, reducing errors and improving quality across more than 330 languages. What it does Translates text and documents using AI. Compares translations from multiple AI models. Provides a consensus translation to improve accuracy. Supports 330+ languages. Helps users evaluate and refine translations. What makes it good Higher accuracy: Compares multiple AI translations instead of relying on a single model. Better quality: Consensus technology helps identify the strongest translation. Fast and easy: Delivers results in seconds with a simple interface. Broad language support: Covers over 330 languages. Useful for professionals and travelers: Suitable for business, marketing, legal, technical, and everyday communication.
Format Factory - Format Factory is a browser-based workbench for common audio and video conversion jobs. It supports batch processing for video and audio conversion, compression, merging, audio extraction, and audio removal, with transparent credits and temporary results. Try Format Factory for practical media conversion without installer bundles or per-file setup.
HappyHorse.AI - HappyHorse.AI is a multimodal AI video generator for creators, marketers, and social media teams. Users can create videos from text prompts, animate images, or generate videos from reference characters. The platform focuses on browser-based video creation and native audio support.
NeatScribe - NeatScribe is designed for anyone who needs to reuse information from recorded speech. It creates timestamped transcripts, supports translation, and exports results as text files, documents, subtitles, captions, or timed lyrics.
Sakana Marlin - Sakana Marlin is Sakana AI's first commercial product: an autonomous 'Ultra Deep Research' agent positioned as a Virtual Chief Strategy Officer. Given a research topic, Marlin works autonomously for up to roughly eight hours, forming hypotheses, gathering and reconciling information from online sources, and synthesizing a detailed strategy report up to ~100 pages plus executive-summary slides. It builds on Sakana AI research including AB-MCTS (NeurIPS 2025 Spotlight) and The AI Scientist (published in Nature). A closed beta ran from April 2026 with ~300 professionals across finance and consulting.
Research Engineer, Language – Generative AI - LLM research and engineering: evaluation, data processing for pretraining/finetuning, efficient training & inference.
Research Software Engineer – Generative AI (AI Frontiers Lab) - Join Microsoft Research AI Frontiers in NY or Redmond; help define next-gen AI platforms.
Senior Deep Learning Algorithm Engineer - Develop algorithms for AI/DL, data analytics, and scientific computing; advance model architectures and distributed training.
Deep Learning Software Engineer, TensorRT Performance (New College Grad 2026) - Integrate TensorRT into open-source frameworks (PyTorch, TensorRT-EdgeLLM); graph compilers and codegen.
Machine Learning Engineer Internship, WebML - Internship working on WebML (browser-native model inference). Remote US.
QUICK HITS
Meta Ordered to Pay $25B to New Mexico Over Cambridge Analytica - Meta ordered to pay $25 billion to New Mexico over the Cambridge Analytica data breach. The largest fine yet in the scandal, highlighting the ongoing legal impact of data privacy issues.
Walmart Launches Agent Gateway for AI Shopping Agents - Walmart launched Agent Gateway, a platform for AI agents to manage transactions, bypassing Amazon's block on AI shopping. Agents can now shop without sharing personal info.
AI Reveals Critical PHP Vulnerabilities, Governments React - AI found critical PHP vulnerabilities, causing government websites to go dark. The NSA advises updates or shutdowns. Node.js also patched four high-severity issues.
Legora Secures $550M Series E at $8.8B Valuation - Legora just raised $550 million at an $8.8 billion valuation, solidifying its place in the legal tech market. Why you care: This shows AI is moving from assistants to full-service productivity tools for legal professionals.
GM Adds AI Assistant to OnStar in 2027 Cadillac Lyriq EV - GM is integrating an AI assistant into OnStar for the 2027 Cadillac Lyriq EV, offering live views and crash recordings. This move aims to stay competitive with Chinese automakers and transform OnStar into a connected intelligence platform.
NSA Warns of Critical PHP Holes in 33M Sites - NSA warns of critical PHP flaws in 33M sites. Update to PHP 8.5.12 ASAP to avoid risks.
THAT'S A WRAP
Our newsletter reaches a diverse audience of tech professionals, investors, engineers, managers, and business leaders worldwide. Get in touch today.
Be the first to comment
Get the daily edition delivered free every morning.
Join 3,565 builders reading daily.
The Browser Company is ending support for its Arc browser on December 22, 2026, directing users to its AI-first browser, Dia. The shutdown is staged, with Arc Search losing AI features on November 23. User reactions on social media are largely disappointed.
Plurilock is set to vote on acquiring Quantum eMotion, aiming to enhance cybersecurity capabilities with quantum tech. Vote date TBA, but the acquisition could drive innovation and competition in the quantum tech market.
A year-old startup has raised a $1 billion Series A, pushing its valuation to $6 billion. The company, founded by three young individuals with impressive backgrounds, aims to build an AGI lab focused on data-efficient models.