Latest Tech News

Stay updated with the latest in technology, AI, cybersecurity, and more

Filtered by: struct Clear Filter

OpenAI Poaches 4 High-Ranking Engineers From Tesla, xAI, and Meta

OpenAI has hired four high-profile engineers away from rivals, including David Lau, former vice president of software engineering at Tesla, to join the company’s scaling team, WIRED has learned. The news came via an internal Slack message sent by OpenAI cofounder Greg Brockman on Tuesday. Lau is joined by Uday Ruddarraju, the former head of infrastructure engineering at xAI and X, Mike Dalton, an infrastructure engineer from xAI, and Angela Fan, an AI researcher from Meta. Both Dalton and Rudda

# [derive(Clone)] Is Broken

use std::sync::Arc; struct NoClone ; struct WrapArc <T>(Arc<T>); fn main () { let foo = WrapArc (Arc:: new (NoClone)); let foo_ = foo. clone (); } Do you think this code should compile? What about the following code: struct AlwaysEq <T>(T); impl <T> PartialEq for AlwaysEq <T> { fn eq (& self , _other: & Self ) -> bool { true } } impl <T> Eq for AlwaysEq <T> {} struct NotEq ; struct WrapAlwaysEq <T>(AlwaysEq<T>); fn assert_is_eq (_: impl Eq ) {} fn main () { let x = WrapAlwaysEq ( AlwaysEq (No

The Download: how AI could improve construction site safety, and our Roundtables conversation with Karen Hao

More than 1,000 construction workers die on the job each year in the US, making it the most dangerous industry for fatal slips, trips, and falls. A new AI tool called Safety AI could help to change that. It analyzes the progress made on a construction site each day, and flags conditions that violate Occupational Safety and Health Administration rules, with what its creator Philip Lorenzo claims is 95% accuracy. Lorenzo says Safety AI is the first one of multiple emerging AI construction sa

Cua (YC X25) is hiring an engineer

Cua is building the infrastructure that lets general AI agents safely and scalably use Computers and Apps like humans do. With 8.9k+ GitHub stars in just 4 months and backing from Y Combinator, we’re providing: An open-source framework for building and evaluating general-purpose AI agents A cloud container platform for sandboxed, scalable agent execution environments A blueprint for what production-grade general agent systems should look like - backed by research We're looking for our first

Automattic puts Tumblr migration to WordPress on hold

In Brief Automattic CEO Matt Mullenweg confirmed that the company is no longer working on migrating its Tumblr blogging platform to WordPress, as previously announced. The exec shared the news on The Verge’s Decoder podcast, published on Monday, saying that the project is “on hold” for an indefinite period. “What we decided is that we want to focus as much on the things that are going to be noticeable to users and that users are asking for,” he told the site. “This was more like an infrastruct

Cua (YC X25) Is Hiring a Founding Engineer

Cua is building the infrastructure that lets general AI agents safely and scalably use Computers and Apps like humans do. With 8.9k+ GitHub stars in just 4 months and backing from Y Combinator, we’re providing: An open-source framework for building and evaluating general-purpose AI agents A cloud container platform for sandboxed, scalable agent execution environments A blueprint for what production-grade general agent systems should look like - backed by research We're looking for our first

The Email Startup Graveyard: Why 80%+ of Email Companies Fail

The Email Startup Graveyard: Why 80%+ of Email Companies Fail While many email startups have invested millions in solving perceived problems, we at Forward Email have focused on building reliable email infrastructure from scratch since 2017. This analysis explores the patterns behind email startup outcomes and the fundamental challenges of email infrastructure. Note Key Insight: Most email startups don't build actual email infrastructure from scratch. Many build on top of existing solutions li

Performance Debugging with LLVM-mca: Simulating the CPU

Some time ago I had a performance problem that wasn’t easy to explain by just looking at the code, since the version I expected to be faster was actually slower. Since the problem is simple yet illustrative, I am using it as a showcase on how to debug performance issues using llvm-mca. According to it’s documentation llvm-mca is a performance analysis tool that uses information available in LLVM (e.g. scheduling models) to statically measure the performance of machine code in a specific CPU. In

Scaling smarter: How enterprise IT teams can right-size their compute for AI

This article is part of VentureBeat’s special issue, “The Real Cost of AI: Performance, Efficiency and ROI at Scale.” Read more from this special issue. AI pilots rarely start with a deep discussion of infrastructure and hardware. But seasoned scalers warn that deploying high-value production workloads will not end happily without strategic, ongoing focus on a key enterprise-grade foundation. Good news: There’s growing recognition by enterprises about the pivotal role infrastructure plays in e

The inference trap: How cloud providers are eating your AI margins

This article is part of VentureBeat’s special issue, “The Real Cost of AI: Performance, Efficiency and ROI at Scale.” Read more from this special issue. AI has become the holy grail of modern companies. Whether it’s customer service or something as niche as pipeline maintenance, organizations in every domain are now implementing AI technologies — from foundation models to VLAs — to make things more efficient. The goal is straightforward: automate tasks to deliver outcomes more efficiently and s

PJ5 TTL CPU

Well, we did it… we cracked 4MIPS on our TTL CPU. I have to admit I’m both relieved and surprised. The key parts to achieving this was to decrease the latency for the instruction decode and also the flag register writing. The time to decode an instruction, perform a bit of math and then write to the flags was at around 270nS. With the changes we made, the instruction decode had a radical reworking and the flag writing had some big changes as well, means we go this down to under 200nS. Which m

Parameterized types in C using the new tag compatibility rule

June 26, 2025 nullprogram.com/blog/2025/06/26/ C23 has a new rule for struct, union, and enum compatibility finally appearing in compilers starting with GCC 15, released this past April, and Clang later this year. The same struct defined in different translation units (TU) has always been compatible — essential to how they work. Until this rule change, each such definition within a TU was a distinct, incompatible type. The new rule says that, ackshually, they are compatible! This unlocks some

Structured Output with LangChain and Llamafile

2 minutes read This article shows how one can teach Llamafile to handle structured outputs like JSON. If you’re already familiar with LangChain, you’ll know that popular models like OpenAI include their own implementations of with_structured_output. Using it is straightforward: All we need is to derive a new class from Pydantic’s BaseModel. The rest happens transparently. You don’t need to teach the LLM anything. Using Llamafile This isn’t currently possible with Llamafile, which I’m using i

How a data-processing problem at Lyft became the basis for Eventual

When Eventual founders Sammy Sidhu and Jay Chia were working as software engineers at Lyft’s autonomous vehicle program, they witnessed a brewing data infrastructure problem — one that would only become larger with the rise of AI. Self-driving cars produce a ton of unstructured data from 3D scans and photos to text and audio. There wasn’t a tool for Lyft engineers that could understand and process all of those different types of data at the same time — and all in one place. This left engineers

How CISOs became the gatekeepers of $309B AI infrastructure spending

Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more Enterprise AI infrastructure spending is expected to reach $309 billion by 2032. The winners won’t be determined by who has the best models; it’ll come down to who controls the infrastructure layer that makes AI operational at scale. Security vendors are making the most aggressive moves. Palo Alto Networks, CrowdStrike and Cisco each repo

How a data processing problem at Lyft became the basis for Eventual

When Eventual founders Sammy Sidhu and Jay Chia were working as software engineers at Lyft’s autonomous vehicle program, they witnessed a brewing data infrastructure problem — and one that would only become larger with the rise of AI. Self-driving cars produce a ton of unstructured data from 3D scans and photos to text and audio. There wasn’t a tool for Lyft engineers that could understand and process all of those different types of data at the same time — and all in one place. This left engine

How fast are Linux pipes anyway?

2022-06-01 How fast are Linux pipes anyway? The challenge, and a slow first version # First of all, let’s start with measuring the performance of the fabled FizzBuzz program, following the rules laid down by the StackOverflow post: % ./fizzbuzz | pv >/dev/null 422GiB 0:00:16 [36.2GiB/s] pv is “pipe viewer”, a handy utility to measure the throughput of data flowing through a pipe. So fizzbuzz is producing output at a rate of 36GiB/s. fizzbuzz writes the output in blocks as big as the L2 cache,

Low-Temperature Additive Manufacturing of Glass

Researchers used the low-temperature additive manufacturing process to build the glass cups above. The optical behavior of the printed cups can be tailored by altering the chemical components of the inks. Additive manufacturing (3D printing) holds promise for fabricating complex glass structures that would be unattainable with traditional glass manufacturing techniques. Lincoln Laboratory’s innovative approach allows additive manufacturing of multimaterial glass items without the need for costl

Foundry (YC F24) Hiring Early Engineer to Build Web Agent Infrastructure

Location: San Francisco, CA Why Foundry Exists: Most of what people do at work sucks—it's manual, repetitive, and wastes time. Recruiters spend hours every day on LinkedIn, CRMs, email, and tedious data entry tasks instead of focusing on the things humans actually do well: building relationships, strategy, and decision-making. We're building tools so that AI agents can use web browsers exactly like humans, navigating enterprise apps like Salesforce, SAP, or Workday without constant manual inte

What I talk about when I talk about IRs

I have a lot of thoughts about the design of compiler intermediate representations (IRs). In this post I’m going to try and communicate some of those ideas and why I think they are important. The overarching idea is being able to make decisions with only local information. That comes in a couple of different flavors. We’ll assume that we’re compiling a method at a time, instead of a something more trace-like (tracing, tracelets, basic block versioning, etc). Control-flow graphs A function wi

Cutting cloud waste at scale: Akamai saves 70% using AI agents orchestrated by kubernetes

Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more Particularly in this dawning era of generative AI, cloud costs are at an all-time high. But that’s not merely because enterprises are using more compute — they’re not using it efficiently. In fact, just this year, enterprises are expected to waste $44.5 billion on unnecessary cloud spending. This is an amplified problem for Akamai Technol

Groq just made Hugging Face way faster — and it’s coming for AWS and Google

Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more Groq, the artificial intelligence inference startup, is making an aggressive play to challenge established cloud providers like Amazon Web Services and Google with two major announcements that could reshape how developers access high-performance AI models. The company announced Monday that it now supports Alibaba’s Qwen3 32B language mode

Flies grow their gyroscopes: Study reveals how flight stabilizers take shape

This article has been reviewed according to Science X's editorial process and policies . Editors have highlighted the following attributes while ensuring the content's credibility: Electron microscopy image showing a haltere developed under normal conditions (left) and a deformed haltere in a genetically modified fruit fly model (right). Credit: Instituto de Neurociencias UMH CSIC A team from the Institute for Neurosciences (IN), a joint center of the Spanish National Research Council (CSIC) a

“Language and Image Minus Cognition”: An Interview with Leif Weatherby

by Robin Manley Leif Weatherby is an Associate Professor of German at New York University, where he directs the Digital Theory Lab. Robin Manley spoke with Dr. Weatherby about his latest book, Language Machines: Cultural AI and the End of Remainder Humanism (University of Minnesota Press, 2025), which argues that Large Language Models (LLMs) have effected a separation of cognition from language and computation in a form that corresponds to earlier structuralist theories. Robin Manley: In the i

Here's why network infrastructure is vital to maximizing your company's AI adoption

Weiquan Lin/Getty Images When companies begin taking the first steps toward AI adoption, one of the first pieces of advice they receive is to address the quality of their data. However, another foundational element that is often overlooked, but is just as critical to the success of AI deployment, is network infrastructure. At Cisco Live, ZDNET spoke with Anurag Dhingra, SVP and GM of the Enterprise Connectivity and Collaboration Group, to learn more about the role network infrastructure plays

“Two years of work in two months”: States cope with Trump broadband overhaul

The Trump administration has upended plans that state governments made to distribute $42 billion in federal broadband funding, forcing state officials to scrap much of the preparation work they did over the previous couple of years. Secretary of Commerce Howard Lutnick essentially put the Broadband Equity, Access, and Deployment (BEAD) program on hold earlier this year and last week announced details of a rules overhaul that requires states to change how they distribute money to Internet servic

Amazon to invest $20 billion in AI data centers across Pennsylvania

What just happened? Amazon Web Services, one of the world's three largest cloud providers, is increasingly focused on artificial intelligence and large language models. The company is now planning a significant expansion of its AI infrastructure initiatives in Pennsylvania. Amazon has announced plans to invest at least $20 billion in new cloud computing facilities across Pennsylvania. The tech giant, founded by Jeff Bezos, says the move reflects its commitment to advancing artificial intelligen

Nvidia makes big play for Europe with infrastructure deals

Jensen Huang, co-founder and CEO of Nvidia Corp., speaks during a news conference in Taipei on May 21, 2025. Cheng / AFP) (Photo by I-HWA CHENG/AFP via Getty Images) Nvidia on Wednesday announced a slew of partnerships with European countries and companies spanning infrastructure to software as it looks to keep itself at the center of the global artificial intelligence story. Chief Executive Jensen Huang on Wednesday continued his tour of Europe with a keynote at Nvidia's GTC event in Paris, F

Microsoft-backed Mistral launches European AI cloud to compete with AWS and Azure

Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more Mistral AI, the French artificial intelligence startup, announced Wednesday a sweeping expansion into AI infrastructure that positions the company as Europe’s answer to American cloud computing giants, while simultaneously unveiling new reasoning models that rival OpenAI’s most advanced systems. The Paris-based company revealed Mistral Co