Search the archive

See the Full Curriculum & Save $625

AMBITION, FUN AND NOT

Startups promise total freedom, but most quickly become rigid, stressful businesses. Cluely's journey from grand ambition to selling meeting software shows how startups often sacrifice bold ideas for corporate reality.

WHY FAIL FAST CULTURE DOESN'T FLY IN MOST PLACES

Companies resist the "fail fast" mentality due to an inherent aversion to failure, but they can embrace accelerated learning by reframing failures as experiments designed to reduce risk. Prioritizing learning, validating assumptions, and continuously measuring value help businesses deliver meaningful products instead of merely shipping features.

MAKING PROGRESS ON CONTROVERSIAL PROBLEMS

When facing controversial team decisions, clarity emerges by understanding the problem stage, documenting context, transparently communicating process steps, and openly questioning assumptions. Proposing a clear, opinionated recommendation, even if imperfect, facilitates productive discussion and collaborative progress.

FIGHTING THE ENSHITTIFICATION OF PRODUCTS: WHY GOOD PRODUCTS GO BAD AND WHAT TO DO ABOUT IT

Enshittification happens when companies prioritize short-term revenue and metrics, sacrificing user trust and product quality. Product teams can counteract it by advocating strong user-focused visions, ethical metrics, and openly challenging harmful trade-offs.

LIES PER SECOND, MEETINGS PER DECISION RATIO, AND OTHER IMPORTANT BIZ METRICS

"Micromorts" quantify risk as a one-in-a-million chance of death, offering perspective on everyday dangers. Creative business metrics like "Lies per Second" humorously spotlight organizational inefficiencies and poor decision-making habits.

MASTERING CHATGPT: ADVANCED TECHNIQUES FOR WORKPLACE COMMUNICATION AND PRODUCTIVITY (42 MINUTE PODCAST)

Hiten Shah, serial founder of companies like Crazy Egg and KISSmetrics, shares how he transforms ChatGPT into a personal AI coach, sales strategist, and productivity tool by customizing it with operating manuals, personality data, and contextual information.

HOW TO COUNTER A FALSE STATEMENT ABOUT YOU

After unintentionally causing offense, avoid defensiveness. Instead, take immediate action to visibly demonstrate your true intentions. This proactive approach rebuilds trust and counters negative perceptions effectively.

LAID-OFF WORKERS SHOULD USE AI TO MANAGE THEIR EMOTIONS, SAYS XBOX EXEC

An Xbox executive faced criticism after suggesting laid-off employees seek emotional support and career advice from AI chatbots, sparking controversy over AI's impact on creative industries amid Microsoft layoffs.

HOW TO ALLOCATE TRAINING BUDGETS

A balanced approach that combines structure and flexibility achieves optimal impact.

BUILDING VERTICAL AI AGENTS FOR THE FUTURE

Develop AI agents tailored to specific industries by embedding workflow context, domain expertise, and data, while designing for AI advancements expected in the next year.

Schedule a demo to learn more

HANDLING UNIQUE INDEXES ON LARGE DATA IN POSTGRESQL

A production deployment failed due to a PostgreSQL limitation on unique index size for large text fields. This limitation is because PostgreSQL stores data in 8KB pages and enforces a maximum index entry size of roughly 2.7KB (1/3 of a page) within B-Tree indexes to optimize performance and page splitting. The solution involves creating a new column to store a hash of the large text field, implementing a trigger to automatically generate the hash on insert or update, and then creating a unique index on this hash column.

UNCOMMON USES OF PYTHON IN COMMONLY USED LIBRARIES

Common Python libraries often use Python in uncommon ways. `super()` in Python base classes is often used for cooperative multiple inheritance, and mixins are strategically applied for modularizing optional features. Relative imports are used for correct package resolution, and the `__init__.py` file doesn't always have the same purpose across packages.

WHEN FIGMA STARTS DESIGNING US

Figma is subtly pushing designers towards an engineering-centric approach through features like Auto Layout and Dev Mode. These features encourage premature optimization and a narrowing of the design space by prioritizing structure and consistency over exploration and spontaneity. This trend can lead to a monoculture in design, where everything feels similar due to shared constraints imposed by the tool.

A NON-ANTHROPOMORPHIZED VIEW OF LLMS

Anthropomorphizing LLMs with concepts like consciousness or ethics is misleading because they are essentially complex mathematical functions generating text. The real challenge lies in quantifying and bounding the probability of LLMs generating undesirable sequences, a task that is difficult because of our inability to define what exactly "undesirable" is. Focusing on LLMs as powerful sequence generators, rather than attributing human-like qualities to them, is necessary for clearer thinking about their impact and potential risks.

MORPH

Morph is a fast AI code edit application with models like morph-v3-fast. It is capable of applying edits at 4,500+ tokens/sec by referencing existing code and offers an alternative to slow, full-file rewrites.

BNA UI

BNA UI provides Expo React Native UI components designed for both iOS and Android. The library includes components, charts, hooks, and themes.

ZERO

Zero is an open-source AI email solution focused on data privacy and self-hosting. It allows users to integrate multiple email providers like Gmail while improving email functionality with AI agents.

ADDING A FEATURE BECAUSE CHATGPT INCORRECTLY THINKS IT EXISTS

This dev noticed an influx of ChatGPT screenshots containing ASCII tablature being uploaded to Soundslice, his sheet music scanning service. He discovered that ChatGPT was falsely advertising that Soundslice supported ASCII tablature import, leading users to believe the feature existed. Faced with this unexpected "market demand" driven by AI misinformation, his team decided to actually build an ASCII tab importer.

I USED O3 TO PROFILE MYSELF FROM MY SAVED POCKET LINKS

This dev profiled themselves using their saved Pocket articles before the service shut down. They used OpenAI's o3 model to analyze their saved URLs. o3 accurately predicted the author's age, location, family size, career, and other details based on the content they had saved.

THE ERA OF EXPLORATION

Future AI progress will depend more on exploration, acquiring new and informative experiences, rather than just scaling parameters. Pretraining LLMs inadvertently pays an "exploration tax" by learning a rich sampling distribution, allowing smaller models to later distill reasoning abilities. Exploration has two key axes: "world sampling," deciding which problems to learn from, and "path sampling," determining how to gather data within those problems.

THE FIRST TIME I WAS ALMOST FIRED FROM APPLE

A newly hired Apple engineer almost got fired for including copyrighted T.S. Eliot poetry in the Color Picker's code as an Easter egg, a mistake that taught him a valuable lesson about intellectual property.

CURSOR'S STEALTH BAIT AND SWITCH: FROM UNLIMITED TO UNUSABLE - MY STORY

A Cursor user writes about a frustrating experience of constant, unannounced downgrades and limitations to their "unlimited" AI coding tool subscription, which ended up being a bait-and-switch scenario.

'I'M BEING PAID TO FIX ISSUES CAUSED BY AI'

Businesses are hiring professionals to fix problems and rewrite generic, faulty content and code generated by AI tools like ChatGPT, often due to unrealistic expectations and a lack of understanding of AI's limitations.

Crush your backlog with Devin, starting from $20.

Or, find out why [6] Ramp, Bilt, and Gumroad are already on board.

TIKTOK'S ‘BAN' PROBLEM COULD END SOON WITH A NEW APP AND A SALE

The Trump administration says it's close to an agreement for a TikTok sale that satisfies the law's requirements, but it involves a new, separate version of the app. The deal still needs approval from the Chinese government. TikTok staff are reportedly working on a new version of the app for release in app stores on September 5. Under the current timeline, the original TikTok app will leave app stores as the new one launches, and the old one will stop working entirely in March 2026.

OPENAI EXPERIMENTS WITH NEW "STUDY TOGETHER" TOOL ON CHATGPT

'Study together' is a new tool in ChatGPT that guides users through the study process step-by-step. It creates a more interactive, guided study environment by posing more questions and breaking down topics, helping users actively learn rather than just receive explanations. It checks answers and adapts by offering follow-ups or clarifying explanations as needed. There is currently no official timeline for the rollout of the feature - a growing number of users are reporting access, indicating a limited A/B test or pre-release trial.

TESLA LAUNCHES OASIS SUPERCHARGER WITH SOLAR FARM AND OFF-GRID BATTERIES

Tesla's Oasis Supercharger, an EV charging station with a solar farm and off-grid batteries, is now open in Lost Hills, California. The project consists of 168 chargers, half of which are currently operational. It is equipped with 11 MW of ground-mounted solar panels and canopies. Spanning 30 acres of land, it has 10 Tesla Megapacks with a total energy storage capacity of 39 MWh. The charging station can be operated off-grid. The rest of the stalls and a lounge will be open later this year.

UK FIRM ACHIEVES TRITIUM BREAKTHROUGH, COULD BOOST NUCLEAR FUSION FUEL SUPPLY

UK-based private commercial fusion company Astral Systems claims to be the first firm to successfully breed tritium, a vital fusion fuel, using a fusion reactor. The achievement addresses a significant hurdle in the development of fusion energy. It clearly demonstrates a potential path to scalable tritium production, which could unlock a wide range of applications and capabilities. The ability to generate tritium within the reactor is crucial - a sustainable fusion energy system needs to produce more fuel than it consumes.

SOFTWARE ENGINEERING WITH LLMS IN 2025: REALITY CHECK

Executives at AI infrastructure companies have made bold claims about the technology, but developers often find these claims to fall spectacularly flat. This article looks at how AI is used within the tech industry to see why founders/CEOs are more bullish than devs about AI tools, how widespread usage is among developers, how much time AI tools really save, and more. Many seasoned software engineers were skeptical about AI tools until very recently - now, most are surprisingly enthusiastic and see these tools as a change that will reshape software development.

HOW TO BECOME PASSIONATE ABOUT DELIVERING SHAREHOLDER VALUE

Many engineers believe their jobs are to write good code, thinking that delivering shareholder value is the company's business, not theirs. Delivering shareholder value means making the company stock price go up and stay up. Shareholder value is part of the software design process - engineers should be connected to the financial goals of the company. Not all engineers are working in situations that can affect shareholder value, but those who work in more functional environments have a real opportunity to shape the financial situation of their companies, for good or ill.

THE CODER ‘VILLAGE' AT THE HEART OF CHINA'S A.I. FRENZY

Liangzhu, a suburb in the southern Chinese city of Hangzhou, is a hot spot for entrepreneurs and tech talent. Lured by the low rents and proximity to tech companies like Alibaba and DeepSeek, people flock to the city to explore their own possibilities. The provincial and local governments started offering subsidies and tax breaks to new companies in the region a decade ago, which helped to incubate hundreds of startups. The region has already birthed tech powerhouses like Alibaba, DeepSeek, NetEase, and Hikvision.

THE COMPANIES BETTING THEY CAN PROFIT FROM GOOGLE SEARCH'S DEMISE

At least a dozen new startups are betting on the rapid demise of Google search and pouring millions of dollars into AI tools. These companies are helping businesses understand how AI chatbots gather information and how to steer them towards brands so that they appear in AI searches. The customer interest is a sign of how AI is transforming search and how companies are trying to get ahead of the changes.

Subscribe now.

OPENCODE

opencode is an AI coding agent built for the terminal.

YOU'RE ALL CTO NOW

Developers using AI are transitioning from writing code to managing agents and projects.

ASK HN: WORTH LEAVING POSITION OVER PUSH TO ADOPT VIBE CODING? (HACKER NEWS THREAD)

Developers should at least stay to learn how the technology works, whether they agree with it or not - it will either work better than expected, or they will be able to inform prospective employers that they have experience working with it later.

GIT EXPERTS SHOULD TRY JUJUTSU

Jujutsu is a force multiplier for those who understand how version control works - it automates the tedious mechanics of history editing, letting developers focus on the what instead of the how.

YET ANOTHER ZIP TRICK

A schizophrenic zip file can be interpreted in different ways by different parsers, resulting in different content.

LEARN TO LOVE THE MOAT OF LOW STATUS

The fear of being temporarily low in social status stops people from living richer lives.

GROK 4 BENCHMARKS LEAK WITH 45% SCORE ON HUMANITY LAST EXAM

Leaked benchmarks show that Grok 4 will be state-of-the-art. References to the model have surfaced in the xAI console. If the benchmarks are real, Grok 4 could outperform leaders like Gemini 2.5 Pro, o3 Pro, and Claude 4 Opus. xAI faces competitive pressure to release Grok 4 before the market shifts again - OpenAI, Google, and Anthropic are rumored to be preparing fresh releases for launch.

CHARACTER AI'S REAL-TIME VIDEO GENERATION

Character.AI's TalkingMachines is a real-time, audio-driven video generation model that creates FaceTime-style animations from a single image and voice input.

ADDING MEMORY TO GEMINI 2.5 CHATBOTS

A guide explaining how to use the Gemini API and the open-source mem0 tool to give Gemini 2.5 chatbots long-term memory. The setup allows bots to recall past interactions, personalize responses, and reduce repetition for more context-aware conversations.

THE AMERICAN DEEPSEEK PROJECT

Meta's recent AI shortcomings have created a vacuum in the open-source AI ecosystem that has largely been filled by Chinese models. If current dynamics continue, the AI world will be split between powerful but expensive, closed-source American models and low-cost, ubiquitous but potentially compromised Chinese models. The US likely has a very small window, roughly the next two years, to counter this trend by investing $100-500 million into an open-source model that is as good as the best closed-source model.