TLDR 2024-06-13

Meta is reportedly looking to cut down on its number of vice-president positions from 300 to about 250. The company's job cuts were guided by principles focused on making Meta an even stronger technology company. These principles included removing multiple layers of management, canceling projects, making every organization leaner, building an optimal ratio of engineers to other roles, and investing in AI and other tools. Meta's capital expenditures on AI and Reality Labs will range between $35 billion and $40 billion by the end of 2024.

Apple and OpenAI's partnership isn't expected to generate meaningful revenue for either party, at least at the beginning. Apple believes that pushing OpenAI's brand and technology to its devices is of equal or greater value than monetary payments so it isn't paying OpenAI as part of the deal. The deal isn't exclusive - Apple is already discussing offering other chatbots as additional options. Apple eventually plans to make money from AI by striking revenue-sharing agreements with AI partners.

A recent tweet from Tesla's official account listed a rundown of the company's accomplishments under Elon Musk's leadership since 2018, when the $55 billion pay package was approved by shareholders. The last item on the list claims that the company has two Optimus robots performing tasks in a factory autonomously. Musk recently said that the Optimus robot might go on sale as soon as next year. It is expected to cost 'less than half of a car'.

Cerebras, a California-based company, has demonstrated that its second-generation wafer-scale engine is significantly faster than the world's faster supercomputer in molecular dynamics calculations. It can also perform sparse large language model inference at one-third of the energy cost of a full model without losing any accuracy. Both achievements are possible due to the interconnects and fast memory access enabled by Cerebras' hardware. Cerebras is looking to extend the applications of its wafer-scale engine to a larger class of problems, including molecular dynamics simulations of biological processes and simulations of airflow around vehicles.

Restate is a platform for building resilient applications that tolerate all infrastructure faults. It enables developers to easily build workflows, event-driven applications, and distribute services in a fault-tolerant manner with durable async/await. Restate is great at building lambda workflows as code, transactional RPC handlers, event processing with Kafka, and more. It supports TypeScript, Java, and Kotlin.

NLUX is a conversational AI JavaScript library that provides a UI for large language models. It makes it super simple to integrate powerful large language models into web apps. NLUX features React components and hooks, LLM adapters, streaming LLM output, and custom renderers.

Competitive Excel is a real thing. The Excel World Championship was recently held at the MGM in Las Vegas. This article takes readers to the front rows of the event. It details what the competition entails and what the event was like. While the event may never be mainstream, it proves that spreadsheeting can be a sport.

Founder liquidity is the practice where founders sell a portion of their shares during a funding round. It allows founders to reduce risk and secure personal financial stability while continuing to build the company with a fresh influx of venture capital. The practice is often kept secret because it undermines the narrative of the founder who is 'all-in'. It might change how startups are perceived and valued if it were widely known that founders could de-risk their financial position while their employees remained all-in.

The vast majority of the revenue comes from OpenAI's products and services - the company has been working to boost revenue by selling services to enterprise customers.

Firefox will start automatically upgrading audio, video, and image subresources from HTTP to HTTPS starting with version 127.

10,900 video game industry employees have been laid off so far in 2024 compared to the prior record of 10,500 in 2023.

This article contains a transcript of an interview with Aidan Gomez, CEO of Cohere, where he discusses Cohere's path to profitability, the benefit of competition in the enterprise space, AI's current abilities, the future of large language models, and more.

Systems powered by Nvidia's Hopper architecture dominated the results of two new tests from MLPerf, an AI benchmarking suite, that compare the fine-tuning of large language models and training of graph neural networks.

Elon Musk responded to Apple's integration of OpenAI ChatGPT on some of its devices by threatening to ban all Apple devices from the premises of his companies.