Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Python continues to lead IEEE Spectrum's 2024 programming language rankings, driven by its dominance in AI and education, while SQL remains highly sought after by employers. Rising stars include TypeScript and Rust, with the latter gaining attention for its memory safety features.
CodeViz is a VSCode extension for visualizing and navigating through your codebase's architecture and functions.
Elasticsearch is again considered Open Source software as Elastic has added the AGPL license as an option alongside ELv2 and SSPL, allowing users to choose the license best suited for their needs.
Thousands of screens and videos of old websites, mobile apps, and software from the 1990s to mid-00s.
An interactive tutorial on prompt engineering for Claude that covers topics from basic prompt structure to advanced techniques for building complex prompts.
Sponsored
☎️ Are phone-based jobs about to go extinct?
NFTs haven't recovered despite crypto values rising. While they're hanging on, the technology is in troubled waters. OpenSea, once the largest marketplace for NFTs, is facing pending litigation from the SEC, a previous unreported 'matter' with the FTC, inbounds from US and international tax authorities, heightened competition, accusations of gender discrimination, and employee attrition. This article tells the story of the marketplace's rise and fall.
OpenAI is reportedly in talks to raise funds at a valuation of more than $100 billion, substantially higher than the company's previous $86 billion valuation. Investors include Thrive Capital and Microsoft. The financing will be OpenAI's biggest outside infusion of capital since Microsoft invested nearly $10 billion in January last year. While OpenAI's revenue eclipsed $3.4 billion early this year, the company is on track to lose nearly $5 billion by year end. It has already burned through $8.5 billion on AI training and staffing.
Workers have been making last-minute upgrades to SpaceX's Starship launch pad in South Texas pretty much every day for the last couple of weeks. If all goes well on the Starship's next test flight, the launch pad's mechanical arms will close together to capture the Super Heavy booster as it descends back to Earth. SpaceX announced that Starship and Super Heavy were ready to fly, pending regulatory approval, on August 8. The FAA is still evaluating SpaceX's proposed flight profile. It is unclear how long approval will take, but the state of the launch pad suggests that the next test is still at least a couple of weeks away.
The competition for AI inference is heating up, particularly in terms of power efficiency. ML Commons has released the results of its latest AI inferencing competition. The round included first-time submissions from teams using AMD Instinct accelerators, the latest Google Trillium accelerators, chips from Untether AI, and Nvidia's new Blackwell chip. This article looks at the results - while Nvidia's chips are still ahead in many categories, chips from other manufacturers are starting to catch up.
Sponsored
In this LinkedIn post, Dan shares 3 key insights he's gained about messaging a developer-focused audience from building ScraperAPI and TLDR. Dan has been actively posting on LinkedIn - follow him for developer marketing content [11].
Using ChatGPT to code can be a real productivity boost. Instead of reading examples on Stack Overflow and adapting them to their particular case, developers can instantly get code tailored to their specific needs. It can make developers more efficient, but they need to remember that it is still a tool. It is important to always test the code it provides to quickly discover any hallucinations.
State is time and time is state. The only way to see the passage of time is by measuring changes in observable state. Changes in state matter in that they create new times. If a state update is purely internal and cannot affect the observable state, then it doesn't advance time. This model can be used to reason about abstract systems.
Nvidia recently reported that its sales and profits had more than doubled during the last quarter. The company also projected that sales in the current quarter will increase by 80% compared to a year ago, exceeding earlier estimates. Nvidia's revenue was $30.04 billion last quarter, and its net income rose to $16.95 billion. It will spend $50 billion on repurchasing its own shares. Nvidia now supplies more than 90% of the chips essential to building AI, but competition is on the rise, with other companies developing custom chips to lessen their dependence on Nvidia.
Telegram CEO Pavel Durov has been indicted in France on charges including web-mastering an online platform in order to enable an illegal transaction in an organized group, complicity in drug trafficking, distribution of child pornography, and providing cryptology services without making required declarations to government officials. Telegram has released a statement saying that it follows the law and industry standards on moderation and that it was absurd to claim that a platform or its owner is responsible for the abuse of that platform. Durov was ordered to post bail of 5 million euros. He is forbidden from leaving the country and must report to police twice a week while the case continues.
Sponsored
Streamline and secure identity, access, devices, and hardware inventory all from one platform. See why 6,000+ IT and security teams
Apple is cutting 100 jobs in its digital services teams - laid off employees have 60 days to find another job in the company.
Codenamed 'Puffin', the headset will achieve its remarkably light weight by offloading both the battery and computing hardware to an external tethered puck small enough to fit in users' pockets.
Midjourney's hardware efforts may be related to its AI models for video and 3D generation.
Solving the alignment problem means avoiding a bad form of AI takeover while building the dangerous kinds of superintelligent AI agents and gaining access to the main benefits of superintelligence from those agents.
Some engineers are worried that staying in one job too long might restrict their growth, but there are ways around that.
Sources confirm that OpenAI's Converge 2 startup accelerator program did indeed take place, but the details about what happened inside the program are scarce.
Sponsored
To build for the new AI era, APIs and developer platforms need to evolve. Join API Summit by Kong (free, virtual) [5] to explore the intersection of AIs and APIs, alongside technical leaders from Kong, Goldman Sachs, Axa France, Maersk, and others.
Midjourney is expanding into hardware with a new team based in San Francisco. The move, hinted at by the hiring of ex-Neuralink and Apple Vision Pro engineer Ahmad Abbas, suggests that the hardware could be related to its ongoing development of AI models for video and 3D generation.
Google is rolling out its new features, Gems and Imagen 3, to Gemini Advanced subscribers. Gems allows users to create custom versions of Gemini tailored to specific tasks, offering premade options like learning coaches and coding partners, while Imagen 3 is Google's latest image generation model, now available to generate detailed and artistic images.
OpenAI is in discussions to raise several billion dollars in a new funding round led by Thrive Capital that would value the company above $100 billion, with Microsoft also expected to participate.
Reward models are generally trained as discriminative classifiers. This work from DeepMind uses a language model's logits for yes/no as the reward signal. It found that by allowing a model to use CoT and ensembling, researchers improved performance by 16%.
Cohere's Aya model was able to dramatically improve its win rate vs baseline models by taking advantage of the difference between oracle model performance and routing synthetic data generation.
Table-Augmented Generation is a new paradigm that combines language models with databases to answer complex natural language questions.
Sponsored
How do you know if your LLM output is any good? Human feedback is the gold standard — but it's expensive, slow, and inconsistent. This webinar by Kolena [14] introduces cutting-edge techniques to fine-tune LLMs according to the preferences of human raters. Join the live
OctFusion is a method for generating 3D shapes with diffusion models that achieves both efficiency and high quality. It can produce 3D shapes at arbitrary resolutions in just 2.5 seconds on a single Nvidia 4090 GPU.
Researchers have developed a new approach to video deblurring by integrating an event camera, which captures motion with microsecond temporal resolution, to enhance the clarity of motion-blurred videos.
Cartesian has announced a number of model and system improvements. It has also released an open hybrid State space model.
Stephen Wolfram emphasizes the importance of incorporating philosophical rigor into AI research, addressing core ethical questions. As AI's influence grows, these questions, traditionally part of philosophy, are increasingly relevant. Wolfram advocates for deeper, classical philosophical thinking in AI development to better navigate its implications for humanity.
Despite general headwinds for startups, AI ventures continue to secure substantial funding. U.S. AI startups have achieved nearly 30 deals over $100M in 2024, with Europe not far behind. Major investments include WAYVE ($1B), Mistral AI (~$1B), Helsing ($484M), Poolside ($400M), DeepL ($320M), H ($220M), and Flo Health ($200M).
Sponsored
AI's impact on HR goes far beyond CV screening. Alan Whitaker, Head of AI at BambooHR, offers some fascinating insights from the front
An open source VLM designed specifically for image upcaptioning.
Boson AI has launched RPBench-Auto, an automated evaluation pipeline to benchmark LLMs' role-playing capabilities inspired by ArenaHard and Alpaca Eval.
NVIDIA's Mistral-NeMo-Minitron 8B, a compressed version of the 12B model, achieves state-of-the-art accuracy while being small enough to run on RTX workstations.
NFT marketplace OpenSea has received a Wells Notice from the SEC, indicating the regulator's interest in pursuing enforcement action against the company. Considering NFTs as securities significantly expands the SEC's regulatory power in the crypto space. OpenSea's CEO has vowed to fight the notice and pledged $5 million to support NFT creators and developers who may face similar scrutiny.
Telegram CEO Pavel Durov has been indicted by a French court on charges related to facilitating illegal activities on the messaging app, including the spread of child pornography and drug trafficking. He must remain in France under judicial supervision and is required to pay a 5 million euro deposit.
Starknet's latest upgrade to version 0.13.2 introduces "parallel execution," enabling the Layer 2 chain to process multiple transactions simultaneously. The update, which aims to improve efficiency and reduce confirmation times, also includes a "block packing" feature to optimize block space usage.
Alt L1 Cardano's most significant upgrade in years, the Chang Hard Fork, will introduce onchain governance mechanisms. The upgrade is scheduled for September 1. It will allow ADA holders to elect representatives and vote on improvement proposals, marking the shift to community governance. There will be a Constitutional Committee, Delegate Representatives, and Stake Pool Operators deciding the chain's future.
The Ethereum Foundation (EF) is preparing to release a new report detailing its spending for 2022 and 2023 to be published before Devcon SEA. Preliminary data shows that about 62% of spending went to external grants and 38% to internal teams, with significant investments in building new institutions to support the Ethereum ecosystem long-term.
Web Proofs, a new technology leveraging zkTLS, could unlock mainstream adoption of crypto by enabling verifiable on-chain data from any web2 source. The innovation could revolutionize crypto incentives like airdrops, expanding their reach beyond the current user base, and facilitate seamless integration of on-chain and off-chain activities, potentially disrupting traditional marketplaces.