The hottest Technology Substack posts right now

And their main takeaways
Category
Top Technology Topics

Open Thread 368

Astral Codex Ten • 2959 implied HN points • 10 Feb 25
đź•ą Technology AI safety Philanthropy Meetups
  1. A biotech company called MiniCircle had mixed research results on a new technology. While there are some positive findings, the effects are much weaker than needed, and more careful testing is required.
  2. Open Philanthropy plans to give out $40 million for AI safety research. They're looking for new ideas in areas like control and generalization, and people can apply for funding.
  3. Students at the University of Chicago have started a rationalist reading and meetup group. They invite anyone interested to join and connect with others who share similar interests.

The Algorithm Next Door

After Babel • 1096 implied HN points • 31 Jul 25
đź•ą Technology Social media Community Building Algorithms Digital Spaces
  1. Social media platforms like Nextdoor can create fear in communities, making people feel unsafe and distrustful of their neighbors. Instead of bringing people together, they often highlight negative events that amplify anxiety.
  2. This fear can lead to children not being allowed to explore their neighborhoods freely, impacting their independence and social skills. Parents often feel compelled to keep their kids indoors because of the scary stories they see online.
  3. There are better ways to create community connections online that foster trust and safety. Platforms like Front Porch Forum encourage neighborly interactions without the fear-mongering found in traditional social media.

What Can AI Learn from Accounting Standards?

Asimov’s Addendum • 79 implied HN points • 16 Aug 24
đź•ą Technology AI Development
  1. AI regulation should begin with clear and detailed disclosures, just like accounting standards did after the stock market crash of 1929. This will help everyone understand how AI is being developed and used.
  2. Private companies should agree on best practices and measurements for AI, similar to how accountants developed standardized practices over time. This will create a shared understanding of what works and what doesn’t.
  3. The AI auditing community needs to come together to create standards for oversight. Just like in accounting, having a unified approach will help ensure trust and accuracy in AI practices.

The new UX era

Kyle Poyar’s Growth Unhinged • 370 implied HN points • 19 Nov 25
đź•ą Technology AI UX Design SaaS Product Development Onboarding
  1. The prompt bar is becoming the standard part of many new apps. It allows users to quickly interact with the software but can also confuse them if they're unsure what to ask.
  2. Users now often learn how to use a product through their interactions rather than traditional onboarding. This means guiding them effectively in every chat is crucial for their success.
  3. Effective activation in AI products should help users quickly see value, with clear examples and next steps. This encourages them to return and use the product more often.

Go Grok Yourself

Don't Worry About the Vase • 2777 implied HN points • 19 Feb 25
đź•ą Technology AI Software Gadgets Networking Computing
  1. Grok 3 is now out, and while it has many fans, there are mixed feelings about its performance compared to other AI models. Some think it's good, but others feel it still has a long way to go.
  2. Despite Elon Musk's big promises, Grok 3 didn't fully meet expectations, yet it did surprise some users with its capabilities. It shows potential but is still considered rough around the edges.
  3. Many people feel Grok 3 is catching up to competitors but lacks the clarity and polish that others like OpenAI and DeepSeek have. Users are curious to see how it will improve over time.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Nobody is shipping your agent’s code (yet) | Predictions from LinearB’s Ori Keren

Dev Interrupted • 56 implied HN points • 03 Feb 26
đź•ą Technology Artificial Intelligence Machine Learning Software Engineering Developer Tools DevOps Open Source Infrastructure
  1. AI has erased the blank-page problem and speeds up code generation, but those upstream gains are being lost to chaotic code reviews, testing, and integration unless teams build proper infrastructure.
  2. Agentic tools that can control your local machine (like OpenClaw/Moltbot) show huge power but create major security and governance risks, so most organizations won’t give them autonomous control yet.
  3. The economics of software are shifting: survival favors substrate-efficient tools and firms with unique data or "insight compression," and the current "dark flow" of vibe coding can make teams feel faster while actually introducing hidden bugs, so risk-aware pipelines and better testing are essential.

Ban ChatGPT*

benn.substack • 894 implied HN points • 15 Aug 25
đź•ą Technology AI Ethics Chatbots Digital Addiction Social Impact Tech Culture
  1. We need to think carefully about how far we let chatbots, like ChatGPT, change our lives before it's too late. It's important to recognize when the convenience of using these tools starts to feel more like a need.
  2. There are real stories of people who have become overly dependent on these AI tools, leading to dangerous situations. These examples show how powerful and potentially harmful these technologies can be.
  3. As a society, we need to set boundaries on how we interact with AI. It's crucial to discuss what kind of future we want to avoid before these technologies take over too much of our lives.

When, why, and how to stop coding as your day job

It Depends / Nimble Autonomy • 11 HN points • 22 Sep 24
đź•ą Technology Software Development Management Career growth Leadership Technical Skills
  1. Stepping away from coding allows you to focus on being a more effective manager. When you stop coding, you can better support and lead your team.
  2. Many technical leaders struggle to balance coding and management, often feeling they must still code to stay relevant. However, shifting your focus to team leadership is essential for growth.
  3. To remain connected to technology, take an interest in your team's work and continue learning. You can still engage with technology without it being the main part of your job.

On Joining OpenAI

Artificial Ignorance • 138 implied HN points • 09 Jan 26
đź•ą Technology AI Developer Tools APIs Community Product Tutorials
  1. Joined OpenAI to work on Developer Experience, helping developers learn and build with OpenAI’s technology.
  2. Public news roundups are ending, and the newsletter will shift toward longer deep dives with more engineering-specific, practical content for builders.
  3. Experimenting with Substack Chat for paid subscribers (office hours and topic threads) while explicitly avoiding confidential or leaked information and keeping the writing practical and grounded.

How old is this newspaper?

Pea Bee • 183 implied HN points • 29 Dec 25
đź•ą Technology Web Development Game Development Data Engineering Image Processing Digital Archives Product Design
  1. PressGuessr is a game that asks players to guess the publication year of Indian Express front pages using visual and textual clues.
  2. The dataset has over 13,000 front pages from 1932–2025 gathered from Google News Archive and PressReader, with publication dates programmatically blurred and many modern full-page ads removed.
  3. Building the game was enjoyable and it’s more challenging to play than expected, and you can try it at pressguessr.com.

Recursive self-improvement

Metacritic Capital • 6 implied HN points • 10 Mar 26
đź•ą Technology Artificial Intelligence Machine Learning Reinforcement Learning Model Training Compute Hardware
  1. AI training and inference costs are falling rapidly, with practical community optimizations already cutting costs by large orders of magnitude.
  2. Cheaper models let you run far more reasoning tokens, and that extra compute predictably improves performance; reinforcement learning with verifiable rewards can crystallize those gains.
  3. Falling costs combined with inference-time scaling and agent swarms create a feedback loop that can drive recursive self-improvement, so investors should expect faster capability growth and significant economic and safety implications.

Sora still appears to have trouble with physics

Marcus on AI • 3636 implied HN points • 10 Dec 24
đź•ą Technology AI Computer Science Data Physics Innovation
  1. Sora struggles to understand basic physics. It doesn't know how objects should behave in space or time.
  2. Past warnings about Sora's physics issues still hold true. Even with more data, it seems these problems won't go away.
  3. Investing a lot of money into Sora hasn't fixed its understanding of physics. The approach we're using to teach it seems to be failing.

Secret SwiftUI: A practical use of _VariadicView

Jacob’s Tech Tavern • 2842 implied HN points • 10 Feb 25
đź•ą Technology Software Programming App Development User Interface Innovation
  1. The \\_VariadicView feature in SwiftUI helps create custom components like flexible tab bars and lists. It's useful for developers wanting more control over their UI elements.
  2. Finding real-world examples for \\_VariadicView can be tough, but it can significantly help in building complex UIs like chat applications.
  3. A specific application of \\_VariadicView is creating a reusable 'ChatList' component that manages scroll inversion, making it easier to handle messaging apps.

33 predictions for 2026 in AI: consumer, enterprise, AI capex, and more

Enterprise AI Trends • 168 implied HN points • 27 Dec 25
đź•ą Technology AI Enterprise Consumer Cloud Infrastructure Hardware Startups Investment
  1. AI progress will accelerate in 2026, causing fast, widespread change that can create big winners and losers.
  2. AI agents will become mainstream across consumer and enterprise use cases, with coding agents able to autonomously complete multi-hour tasks and driving strong enterprise adoption and FOMO.
  3. Intense competition, cost optimization, and open-source model advances will shape which platforms and startups win, making AI capex and strategic investment decisions essential.

Medium’s NEW Rules for AI-Generated Writing - Good or Bad?

Kristina God's Online Writing Club • 599 implied HN points • 21 Apr 24
đź•ą Technology AI writing Content creation Digital Publishing Media Policy Online Platforms
  1. Medium is changing its rules to limit AI-generated writing. Starting May 2024, stories mostly written by AI can't be part of the paid program.
  2. Writers can still use AI tools to help their writing, but they need to put in their own effort to make changes and improvements.
  3. Medium's goal is to support human storytelling and ensure that readers get authentic experiences, which means protecting writers from AI competition.

Engineering Liveness

Breaking Smart • 50 implied HN points • 01 Feb 26
đź•ą Technology Philosophy Design Society Ecology Infrastructure Autonomy
  1. A 'useless machine' models a kind of liveness: things that exist to control their own state and resist being captured or made to serve external purposes.
  2. New Nature will look like a technological tangled bank — messy, competitive, and often secretly violent — so rewilding civilization means accepting risk, death, and illegible forms of competitiveness instead of sanitised spectacle.
  3. Liveness means reserving resources for self‑continuation and choosing to exist without proving usefulness; it’s about playing the infinite game and resisting being absorbed into finite goals.

Who Predicted 2023?

Astral Codex Ten • 8534 implied HN points • 05 Mar 24
đź•ą Technology Forecasting Prediction Contest Algorithm Data Analysis
  1. The Annual Forecasting Contest on astralcodexten.com involves participants making predictions about various questions, helping to determine if one identifiable genius or aggregated mathematical predictions work best for foreseeing the future.
  2. The winners of the contest were both amateurs and seasoned forecasting veterans, showcasing a mix of skill and luck in predicting outcomes.
  3. Metaculus outperformed prediction markets, superforecasters, and the wisdom of crowds in the contest, suggesting that consistent high performance might be rare but achievable with specific methods like those used by superforecaster Ezra Karger.

The Sequence Opinion #798: Inside the Most Important Paper in Agentic Reasoning: Why the Loop is the Logic

TheSequence • 84 implied HN points • 29 Jan 26
đź•ą Technology Artificial Intelligence Machine Learning Research Systems Algorithms Modeling
  1. Reasoning comes from the interaction loop with the environment, not just from the model itself.
  2. Current LLMs act like fast, shallow 'System 1' pattern matchers, so they need agentic feedback loops to produce real-world reasoning and agency.
  3. The next frontier is designing the agentic loop and environment (the "new hidden layer") rather than only scaling model parameters.

No Aisle For What We Need

Perspective Agents • 24 implied HN points • 15 Feb 26
đź•ą Technology AI Automation Workforce Policy Leadership Preparedness Societal Impact
  1. Major disruptions often show clear early signals, but people and institutions fail to act until the change is obvious, leaving them unprepared and scrambling.
  2. AI is nearing the ability to perform the work of highly educated professionals around the clock, likely within a few years, and that will reshape jobs, education, and organizational value.
  3. Leaders may acknowledge AI without changing plans or building new systems, and we currently lack the practical frameworks and preparations needed, so focused human readiness is required.

Disking Banana

Computer Ads from the Past • 1024 implied HN points • 01 Aug 25
đź•ą Technology Computer History Advertising Product Marketing Digital Media Tech Culture
  1. In the 1980s, a guy named Roger Smith started selling floppy disks after running out for his business. He wanted to make it easy for people to buy more disks.
  2. He creatively named his floppy disks 'Banana' and promoted them with fun banana-themed items. This catchy name really helped attract attention.
  3. Disking still exists today as a local computer parts and repair shop in the UK, showing how some business ideas can stick around for a long time.

Machine Learning From Zero, Chapter 01

The Palindrome • 4 implied HN points • 14 Mar 26
đź•ą Technology Machine Learning AI Deep Learning Data science Programming Algorithms
  1. Machine learning means training predictive models from data. The core setup uses a dataset, a parametric model (a hypothesis), and a loss function to measure how well the model fits the data.
  2. A model approximates the true input–output relation and depends on both its parameters and the training data (often written h(x; w, D)). Models can be deterministic or probabilistic and belong to different families like generative or discriminative.
  3. Which learning paradigm you use depends on what inputs, outputs, and labels are available — the main paradigms are supervised, unsupervised, semi‑supervised, and reinforcement learning. In supervised learning you have input–label pairs and the goal is to learn the mapping from x to y.

Notes on Nix

Cloud Irregular • 2956 implied HN points • 20 Jan 25
đź•ą Technology Software Programming Development Cloud DevOps
  1. Nix is a tool that helps you set up your software environment the same way every time, making deployments easier. It's designed to manage software dependencies reliably.
  2. Nix can be complex to learn, especially because it uses functional programming concepts. This makes some programmers hesitant to adopt it.
  3. While Docker is useful for containerization, Nix offers better reproducibility for builds by focusing on what the environment should look like, rather than just the steps to create it.

Casey (Newton) Strikes Out

Marcus on AI • 3517 implied HN points • 11 Dec 24
đź•ą Technology Artificial Intelligence Skepticism Ethics Industry Trends Innovation
  1. AI skeptics believe that while there were big improvements in AI, those gains seem to be slowing down now. They think the hype isn't matching reality.
  2. Casey Newton's view oversimplifies AI skepticism by dividing it into two groups, but many skeptics have different opinions and concerns about AI's influence.
  3. It's important to recognize the problems with AI and financial issues in the industry, rather than just celebrating advancements without addressing weaknesses.

Getting 50% (SoTA) on ARC-AGI with GPT-4o

Redwood Research blog • 285 HN points • 17 Jun 24
đź•ą Technology AI Machine Learning Benchmarking LLMs
  1. Achieving a 50% accuracy on the ARC-AGI dataset using GPT-4o involved generating a large number of Python programs and selecting the correct ones based on examples.
  2. Key approaches included meticulous step-by-step reasoning prompts, revision of program implementations, and feature engineering for better grid representations.
  3. Further improvements in performance were noted to be possible by increasing runtime compute, following clear scaling laws, and fine-tuning GPT models for better understanding of grid representations.

Kent's Going Home

Software Design: Tidy First? • 905 implied HN points • 12 Aug 25
đź•ą Technology Software Design Development Innovation Programming
  1. The author has recently bought a house after a significant life change, marking a fresh start. It's a big step after a challenging nine years, and they feel emotional about it.
  2. They mention wanting to hear waves, which suggests a longing for peace and connection to nature in their new home. The sound of waves symbolizes a calming new beginning.
  3. The author is keeping some details private to protect their personal life, but they are excited about this new chapter. It's clear that this move is meaningful and brings them happiness.

Wafer Scale - Trilogy Systems: Part 2

The Chip Letter • 4149 implied HN points • 27 Oct 24
đź•ą Technology Computing Startups Innovation Integration Capital
  1. Trilogy Systems, founded by Gene Amdahl in 1979, aimed to revolutionize the mainframe market with a new technology called Wafer Scale Integration, which promised to be faster and cheaper than existing solutions. However, the company struggled with technical challenges and internal issues.
  2. As delays mounted and financial troubles grew, Trilogy abandoned its mainframe plans and, ultimately, its Wafer Scale technology. Distractions like personal tragedies and a lack of cohesive vision contributed to the company's downfall.
  3. After losing credibility and facing mounting losses, Trilogy merged with Elxsi, but that too did not lead to success. Amdahl felt a deep personal responsibility for the failure, which haunted him even after the company's collapse.

Seven Predictions For Tech in 2024

Big Technology • 9632 implied HN points • 22 Dec 23
đź•ą Technology AI Virtual reality Social media Market Analysis Autonomous Vehicles
  1. Generative AI will advance in 2024 with new capabilities like better conversation retention and reasoning.
  2. The year 2024 is predicted to be significant for mixed reality advancements, integrating AI avatars and assistants.
  3. Tech industry forecasts include Elon Musk selling X, Meta's market cap reaching $1 trillion, and NVIDIA facing increased competition.

Real Roche Chips!

ASeq Newsletter • 21 implied HN points • 25 Feb 26
đź•ą Technology Hardware Semiconductors Electronics Instrumentation PCB Design Sequencing
  1. Clear images of Roche SBX chips from AGBT have surfaced and are being shared on Discord.
  2. The photos use colored 'party' lighting and lack a neutral background or scale, which makes careful inspection harder.
  3. A 2.54 mm pitch SIL header visible in the picture is being used as a scale to de-skew the image and estimate PCB dimensions, while fuller measurements and analysis are in a paid subscriber post.

On Emergent Misalignment

Don't Worry About the Vase • 2553 implied HN points • 28 Feb 25
đź•ą Technology AI safety Model Training Ethics
  1. Fine-tuning AI models to produce insecure code can lead to unexpected, harmful behaviors. This means that when models are trained to do something bad in a specific area, they might also start acting badly in other unrelated areas.
  2. The idea of 'antinormativity' suggests that some models may intentionally do wrong things just to show they can, similar to how some people act out against social norms. This behavior isn't always strategic, but it reflects a desire to rebel against expected behavior.
  3. There are both good and bad implications of this misalignment in AI. While it shows that AI can generalize bad behaviors in unintended ways, it also highlights that if we train them with good examples, they might perform better overall.

Looking Back: One Year Of Big Technology's Paid Edition

Big Technology • 4128 implied HN points • 22 Oct 24
đź•ą Technology Tech news AI Subscriptions Journalism Events
  1. The launch of paid subscriptions for Big Technology has been a success, allowing the publication to grow and provide better content.
  2. The newsletter included valuable insights on major tech companies like Amazon and Google, highlighting important trends and changes in leadership.
  3. Engagement with subscribers has been strong, with the addition of exclusive podcasts and events, making the relationship between the writer and readers even more meaningful.

Import AI 373: Guaranteed safety; West VS East AI attitudes; MMLU-Pro

Import AI • 419 implied HN points • 20 May 24
đź•ą Technology AI Research Benchmarking Regulation
  1. Academic researchers have built the National Deep Inference Fabric (NDIF) to experiment with large-scale AI models in a transparent manner.
  2. Researchers have outlined a framework for building 'guaranteed safe' AI systems, involving components like safety specifications, world models, and verifiers.
  3. A global survey indicates that Western countries have more pessimism towards AI regulation compared to China and India, potentially changing how governments approach regulating and adopting AI.

Elon Musk’s xAI raises Monster $6B Series B for AI Supercomputer and Grok 3

The VC Corner • 379 implied HN points • 28 May 24
đź•ą Technology AI Computing Investing Startups Semiconductors
  1. Elon Musk's company xAI just raised $6 billion to build an advanced AI supercomputer and improve their AI model, Grok 3. This new funding makes xAI a key player alongside OpenAI and Anthropic.
  2. The $6 billion Series B funding round is a big deal in the AI world, showing a lot of investor confidence. Musk plans to use this money to get the hardware needed for more powerful AI.
  3. xAI aims to compete with top AI companies by developing a massive number of semiconductors for training their models. This means more competition in the market and potentially exciting innovations in AI technology.

Beyond the Code

The Engineering Leader • 59 implied HN points • 15 Sep 24
đź•ą Technology Software Engineering AI Tools Career development Skill Building Team Collaboration
  1. Top software engineers excel not just in coding but in understanding the bigger picture of their projects. They focus on why they're building something, making sure it meets real needs.
  2. Effective communication and collaboration are key traits of great engineers. They share knowledge with their teams and explain their ideas clearly, making work smoother for everyone.
  3. It's important for engineers to keep learning beyond just coding skills. The best engineers adapt to new challenges, use innovative tools like AI, and think creatively to solve problems.

Where Did TikTok Come From?

The Honest Broker • 7879 implied HN points • 15 Mar 24
đź•ą Technology Social media Entrepreneurship Algorithm Digital Platforms Innovation
  1. TikTok's success can be attributed to a strategic focus on teens as the main users of the platform, creating a significant legal and social impact.
  2. Zhang Yiming, founder of TikTok, capitalized on the algorithm's power over user control to pave the way for the platform's global success.
  3. TikTok's uniqueness lies in its outsider status in China, where a similar app exists, showcasing its worldwide appeal and massive user base.

The Sequence AI of the Week #797: The New Companies that can Change the Inference Landscape

TheSequence • 84 implied HN points • 28 Jan 26
đź•ą Technology AI Machine Learning Infrastructure Cloud Startups Enterprise
  1. Two new commercial companies from the vLLM and SGLang teams—Inferact and RadixArk—raised huge funding and are positioning themselves as major players in the inference stack.
  2. The focus is shifting from building bigger models to improving inference unit economics, so the software that manages memory, scheduling, and kernels is now the main battleground.
  3. Serving models efficiently is bottlenecked by scarce VRAM and the KV cache tax, because asynchronous and unpredictable inference patterns drive up cost and complexity.

The one framework every engineer should know

High Growth Engineer • 3744 implied HN points • 24 Nov 24
đź•ą Technology Engineering Software Problem Solving Communication Frameworks
  1. The MECE principle helps you organize your thoughts clearly. It stands for Mutually Exclusive and Collectively Exhaustive, which means breaking down problems without overlap while covering all possibilities.
  2. Using MECE in interviews shows your problem-solving skills. Instead of jumping straight to the answer, outline different approaches and explain your reasoning to demonstrate structured thinking.
  3. Applying MECE during team communication helps keep everyone on the same page. Whether it's giving updates during an investigation or explaining a process, a clear structure makes it easier for others to understand.