Investors on what they really want to hear in a founder's pitch.

“I think that 2025 was the year where the algorithm completely took over, so followings stopped mattering entirely,” LTK CEO Amber Venz Box said.

AI’s early-2025 spending spree featured massive raises and trillion-dollar infrastructure promises. By year’s end, hype gave way to a vibe check, with growing scrutiny over sustainability, safety, and business models.

Plaud Note Pro is a $179 notetaker, which is an excellent recording device first

The electrical grid is facing unprecedented stress from the addition of new data centers. Software could offer a cost-effective way to boost reliability and capacity.

The 2025 Typed Python Survey, conducted by contributors from JetBrains, Meta, and the broader Python typing community, offers a comprehensive look at the current state of Python’s type system and developer tooling. With 1,241 responses (a 15% increase from last year), the survey captures the evolving sentiment, challenges, and opportunities around Python typing in the [...] <a class="btn btn-secondary understrap-read-more-link" href="https://engineering.fb.com/2025/12/22/developer-tool

Incident investigation can be a daunting task in today’s digital landscape, where large-scale systems comprise numerous interconnected components and dependencies DrP is a root cause analysis (RCA) platform, designed by Meta, to programmatically automate the investigation process, significantly reducing the mean time to resolve (MTTR) for incidents and alleviating on-call toil Today, DrP is used [...] <a class="btn btn-secondary understrap-read-more-link" href="https://engineering.fb.c

We’re going behind the scenes of the Meta Ray-Ban Display, Meta’s most advanced AI glasses yet. In a previous episode we met the team behind the Meta Neural Band, the EMG wristband packaged with the Ray-Ban Display. Now we’re delving into the glasses themselves. Kenan and Emanuel, from Meta’s Wearables org, join Pascal Hartig on [...] <a class="btn btn-secondary understrap-read-more-link" href="https://engineering.fb.com/2025/12/17/virtual-reality/meta-ray-ban-display-from-zero-to-poli

Meta’s secure-by-default frameworks wrap potentially unsafe OS and third-party functions, making security the default while preserving developer speed and usability. These frameworks are designed to closely mirror existing APIs, rely on public and stable interfaces, and maximize developer adoption by minimizing friction and complexity. Generative AI and automation accelerate the adoption of secure frameworks at [...] <a class="btn btn-secondary understrap-read-more-link" href="https://

We’re introducing Zoomer, Meta’s comprehensive, automated debugging and optimization platform for AI. Zoomer works across all of our training and inference workloads at Meta and provides deep performance insights that enable energy savings, workflow acceleration, and efficiency gains in our AI infrastructure. Zoomer has delivered training time reductions, and significant QPS improvements, making it the [...] <a class="btn btn-secondary understrap-read-more-link" href="https://enginee

The system we built to ensure our AI agents produce predictable, trustworthy code. The post Background Coding Agents: Predictable Results Through Strong Feedback Loops (Part 3) appeared first on Spotify Engineering.

We explore context engineering for background coding agents and what makes a good migration prompt. The post Background Coding Agents: Context Engineering (Part 2) appeared first on Spotify Engineering.

Shuffle has always been one of Spotify’s most-used features, and also one of the most misunderstood. For... The post Shuffle: Making Random Feel More Human appeared first on Spotify Engineering.

Thousands of merged AI-generated pull requests and the future of large-scale software maintenance. The post 1,500+ PRs Later: Spotify’s Journey with Our Background Coding Agent (Part 1) appeared first on Spotify Engineering.

TL;DR Spotify’s experimentation platform, Confidence, scaled product decision-making across hundreds of... The post Beyond Winning: Spotify’s Experiments with Learning Framework appeared first on Spotify Engineering.

Gitanjali Venkatraman does wonderful illustrations of complex subjects (which is why I was so happy to work with her on our Expert Generalists article). She has now published the latest in her series of illustrated guides: tackling the complex topic of <a href="https://www.thoughtworks.com/content/dam/thoughtworks/documents/blog/mainframe_modernisation_illust

If you’re a regular reader of my site, you’ll have noticed that in the last few months I’ve been making a number of “fragments” posts. Such a post is a short post with a bunch of little, unconnected segments. These are usually a reference to something I’ve found on the web, sometimes a small thought of my own. A few years ago, I wouldn’t have covere

Why does AI write like… that (NYT, gift link). Sam Kriss delves into the quiet hum of AI writing. AI’s work is not compelling prose: it’s phantom text, ghostly scribblings, a spectre woven into our communal tapestry. ❄ ❄ ❄ ❄ ❄ <a href="https://coding-is-like-cooking.info/2025/12/test-desidera

Rob Bowley summarizes a study from Carnegie Mellon looking on the impact of AI on a bunch of open-source software projects. Like any such study, we shouldn’t take its results as definitive, but there seems enough there to make it a handy data point. The key point is that the AI code probably reduced the quality of the code base - at least if static code analysis can be trusted to determ

I’ve been on the road in Europe for the last couple of weeks, and while I was there Thoughtworks released volume 33 of our Technology Radar. Again it’s dominated by the AI wave, with lots of blips capturing our explorations of how to use LLMs and similar technology. “Agents” are the big thing these days but we’re also seeing grow

The best model in the world for coding, agents, and computer use, with meaningful improvements to everyday tasks like slides and spreadsheets.

Claude Sonnet 4.5 sets new benchmark records in coding, reasoning, and computer use while being Anthropic's most aligned model.

Claude Haiku 4.5 matches state-of-the-art coding capabilities from months ago while delivering unprecedented speed and cost-efficiency for complex tasks.

Anthropic raised $13 billion in a Series F round at a $183 billion valuation to expand enterprise offerings, safety research, and international growth.

Anthropic's response to the White House AI Action Plan supports infrastructure and safety measures while calling for stronger export controls.

Parallel agents, Git worktrees, multi-pane Cascade, dedicated terminal, and SWE-1.5 Free

GPT-5.2 is now live in Windsurf! Available for 0x credits for a limited time (paid and trial users). The version bump undersells the jump in intelligence: Biggest leap for GPT models in agentic coding since GPT-5, SOTA coding model at its price point, Default in Windsurf

The most capable model in Windsurf yet, now available at Sonnet prices for a limited time

GPT 5.1, GPT 5.1-Codex, and GPT-5.1-Codex Mini deliver a solid upgrade for agentic coding with variable thinking and improved steerability

SWE-1.5 is our latest frontier model, delivering near-SOTA coding performance at unprecedented speed.

We're partnering with ecosystem vendors who have built hooks support with Cursor.

Graphite has entered into a definitive agreement to be acquired by Cursor.

Bringing design and engineering closer together.

Debug Mode helps you reproduce and fix the most tricky bugs.

How we updated our agent harness to support GPT-5.1-Codex-Max.

More than one million customers around the world now use OpenAI to empower their teams and unlock new opportunities. This post highlights how companies like PayPal, Virgin Atlantic, BBVA, Cisco, Moderna, and Canva are transforming the way work gets done with AI.

OpenAI is strengthening ChatGPT Atlas against prompt injection attacks using automated red teaming trained with reinforcement learning. This proactive discover-and-patch loop helps identify novel exploits early and harden the browser agent’s defenses as AI becomes more agentic.

OpenAI introduces a new framework and evaluation suite for chain-of-thought monitorability, covering 13 evaluations across 24 environments. Our findings show that monitoring a model’s internal reasoning is far more effective than monitoring outputs alone, offering a promising path toward scalable control as AI systems grow more capable.

OpenAI and the U.S. Department of Energy have signed a memorandum of understanding to deepen collaboration on AI and advanced computing in support of scientific discovery. The agreement builds on ongoing work with national laboratories and helps establish a framework for applying AI to high-impact research across the DOE ecosystem.

OpenAI shares new AI literacy resources to help teens and parents use ChatGPT thoughtfully, safely, and with confidence. The guides include expert-vetted tips for responsible use, critical thinking, healthy boundaries, and supporting teens through emotional or sensitive topics.

Google 2025 recap: Research breakthroughs of the year

Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.

Open interpretability tools for language models are now available across the entire Gemma 3 family with the release of Gemma Scope 2.

Google DeepMind and UK AI Security Institute (AISI) strengthen collaboration on critical AI safety and security research

Agents still face challenges working across many context windows. We looked to human engineers for inspiration in creating a more effective harness for long-running agents.

Introducing advanced tool use on the Claude Developer Platform

Code execution with MCP: Building more efficient agents

Beyond permission prompts: making Claude Code more secure and autonomous

Equipping agents for the real world with Agent Skills

NVIDIA will join the U.S. Department of Energy’s (DOE) Genesis Mission as a private industry partner to keep U.S. AI both the leader and the standard in technology around the world. The Genesis Mission, which is part of an Executive Order recently signed by President Trump, aims to redefine American leadership in AI across three Read Article </span

The NVIDIA RTX PRO 5000 72GB Blackwell GPU is now generally available, bringing robust agentic and generative AI capabilities powered by the NVIDIA Blackwell architecture to more desktops and professionals across the world.

Step out of the vault and into the future of gaming with Fallout: New Vegas streaming on GeForce NOW, just in time to celebrate the newest season of the hit Amazon TV show Fallout. To mark the occasion, GeForce NOW members can claim Fallout 3 and Fallout 4 as special rewards, completing a wasteland-ready trilogy Read Article

Physical AI is moving from research labs into the real world, powering intelligent robots and autonomous vehicles (AVs) — such as robotaxis — that must reliably sense, reason and act amid unpredictable conditions.

The Hao AI Lab research team at the University of California San Diego — at the forefront of pioneering AI model innovation — recently received an NVIDIA DGX B200 system to elevate their critical work in large language model inference. Many LLM inference platforms in production today, such as NVIDIA Dynamo, use research concepts that Read Article

AI co-scientists are emerging as a tool to assist human researchers in achieving their research goals. A crucial feature of these AI co-scientists is the ability to generate a research plan given a set of aims and constraints. The plan may be used by researchers for brainstorming, or may even be implemented after further refinement. However, language models currently struggle to generate research plans that follow all constraints and implicit requirements. In this work, we study how to leverage

Identifying specific and often complex behaviors from large language models (LLMs) in conversational settings is crucial for their evaluation. Recent work proposes novel techniques to find natural language prompts that induce specific behaviors from a target model, yet they are mainly studied in single-turn settings. In this work, we study behavior elicitation in the context of multi-turn conversations. We first offer an analytical framework that categorizes existing methods into three families

Diffusion language models (DLMs) enable parallel, order-agnostic generation with iterative refinement, offering a flexible alternative to autoregressive large language models (LLMs). However, adapting reinforcement learning (RL) fine-tuning to DLMs remains an open challenge because of the intractable likelihood. Pioneering work such as diffu-GRPO estimated token-level likelihoods via one-step unmasking. While computationally efficient, this approach is severely biased. A more principled foundati

We introduce Iterated Bellman Calibration, a simple, model-agnostic, post-hoc procedure for calibrating off-policy value predictions in infinite-horizon Markov decision processes. Bellman calibration requires that states with similar predicted long-term returns exhibit one-step returns consistent with the Bellman equation under the target policy. We adapt classical histogram and isotonic calibration to the dynamic, counterfactual setting by repeatedly regressing fitted Bellman targets onto a mod

Recent findings by Cohen et al., 2021, demonstrate that when training neural networks using full-batch gradient descent with a step size of $η$, the largest eigenvalue $λ_{\max}$ of the full-batch Hessian consistently stabilizes around $2/η$. These results have significant implications for convergence and generalization. This, however, is not the case for mini-batch optimization algorithms, limiting the broader applicabilityof the consequences of these findings. We show mini-batch Stochastic Gra

Can you believe it? We’re nearly at the end of 2025. And what a year it’s been! From re:Invent recap events, to AWS Summits, AWS Innovate, AWS re:Inforce, Community Days, and DevDays and, recently, adding that cherry on the cake, re:Invent 2025, we have lived through a year filled with exciting moments and technology advancements […]

The week after AWS re:Invent builds on the excitement and energy of the event and is a good time to learn more and understand how the recent announcements can help you solve your challenges and unlock new opportunities. As usual, we have you covered with our top announcements of AWS re:Invent 2025 that you can […]

Amazon Bedrock now supports reinforcement fine-tuning delivering 66% accuracy gains on average over base models.

Accelerate AI model development with new training features that enable rapid recovery from failures and automatic scaling based on resource availability.

Accelerate AI model development with new training features that enable instant recovery from failures and automatic scaling based on resource availability.

Physical data center maintenance is risky on a global network. We built a maintenance scheduler on Workers to safely plan disruptive operations, while solving scaling challenges by viewing the state of our infrastructure through a graph interface on top of multiple data sources and metrics pipelines.

We have declared “Code Orange: Fail Small” to focus everyone at Cloudflare on a set of high-priority workstreams with one simple goal: ensure that the cause of our last two global outages never happens again.

Cloudflare's H1 2025 Transparency Report is here. We discuss our principles on content blocking and our innovative approach to combating unauthorized streaming and copyright abuse.

Cloudflare’s R2 SQL, a distributed query engine, now supports aggregations. Explore how we built distributed GROUP BY execution, using scatter-gather and shuffling strategies to run analytics directly over your R2 Data Catalog.

We present our 6th annual review of Internet trends and patterns observed across the globe, revealing the disruptions, advances and metrics that defined 2025.