LLMs reward expertise

https://www.seangoedecke.com/og-image.jpg
Using large language models like LLMs requires domain knowledge to get the most out of them. A skilled prompter like Terence Tao can steer the model with expertise in the domain, but this skill is not easily replicable by following tips.

Ten advances in mathematics and theoretical computer science

Devtools must be open source

https://blog.exe.dev/static/og-card.png
Software engineers used to rarely write personal software, but now agents can easily manage customizing software, making it easier to personalize and maintain. Agents can build and manage custom software with minimal programming, allowing for more efficient and flexible use of software.

Windows XP 2002 for the Itanium: Unbridled rage

Ask HN: Who is hiring? (August 2026)

Bucket Robotics: Senior Fullstack Engineer, San Francisco, CA (On-site) - Own massive surface area, build customer-facing app, scale internal tooling, and design data pipelines. Product-minded engineer with systems thinking and curiosity.

Smaller, faster, safer: running Kimi and GLM at scale

https://blog.cloudflare.com/_image?href=https%3A%2F%2Fblog.cloudflare.com%2F_emdash%2Fapi%2Fmedia%2Ffile%2F01KZ1PVA5SC9J87PS1R2B97N1F.png&w=1999&h=1125&f=webp&fit=cover&position=center
Workers AI optimizes large models on GPUs by quantizing the KV cache to FP8, compressing model weights to INT4, and protecting the cache with integrity checks. These techniques enable efficient serving of models like Kimi K-series and GLM with no change in accuracy, supporting more customers at lower costs.

200 Milliseconds

https://200ms.thenodebook.com/og.png
See what really happens in the ~200ms after you press Enter: a single HTTP request traced in real time through DNS, TCP, TLS, the Linux kernel, Node's event loop, and Postgres — then back to the pixel. Scroll distance equals time.

Ask HN: Who wants to be hired? (August 2026)

Several professionals with AI, data, and software engineering backgrounds are seeking roles in various locations, including Brazil, Canada, India, Russia, and the US, with expertise in technologies like Python, SQL, Generative AI, LLMs, and Rust. They are looking for opportunities as Senior AI Engineers, Applied AI, Search/Relevance, RAG, Data/AI Lead, and AI Solutions roles, with some open ...

Celebrating 45 Years of Kermit with the First New C-Kermit Release in 15 Years

https://changelog.complete.org/.within.website/x/cmd/anubis/static/img/pensive.webp?cacheBuster=1.22.0
Please wait a moment while we ensure the security of your connection.

MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video

https://substackcdn.com/image/fetch/$s_!59Bc!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F98d2552b-d84c-4385-9602-bf344dca2ebe_1000x727.png
MiniMax H3 is a next-generation open-weights video model that generates 2K video with real stereo sound up to 15 seconds. It combines text, images, video, and audio inputs with a prompt to create a single video clip, reducing five tasks into one model.

Battle of the Beams

https://upload.wikimedia.org/wikipedia/commons/thumb/9/93/Map_of_Knickebein_transmitters.svg/250px-Map_of_Knickebein_transmitters.svg.png
The Battle of the Beams was a period early in the Second World War when bombers of the German Air Force (Luftwaffe) used a number of increasingly accurate systems of radio navigation for night bombing in the United Kingdom. British scientific intelligence at the Air Ministry fought back with a variety of their own increasingly effective means, involving jamming and deception signals. The ...

Andy Pavlo joins ClickHouse to establish ClickHouse Labs

https://clickhouse.com/_next/image?url=%2Fuploads%2Fandy_pavlo_clickhouse_b978ca187a.png&w=750&q=75
I'm joining ClickHouse to lead ClickHouse Labs, a research team focused on advancing database technology. Our goal is to conduct impactful research and transform ideas into technology that benefits users, building on ClickHouse's strong foundation and exploring emerging AI and agentic technologies.

The Dunning-Kruger effect may just be a data artefact (2020)

https://www.mcgill.ca/oss/files/oss/figure_1_3.png
The Dunning-Kruger effect may not be a real bias in human thinking as it can be replicated with random data and is likely due to measurement error and unreliability. The effect was originally described as people overestimating their competence in areas they are not skilled at, but this may be an artefact of how the data was measured.

Launch HN: Hoplite (YC S26) – Effortlessly deploy cloud coding agents

https://hoplite.sh/opengraph.png
Effortless cloud coding agents that feel good to use.

ZX Spectrum System Tour: Text Mode

https://bumbershootsoft.wordpress.com/wp-content/uploads/2026/05/spectrum_manic_mechanic.png?w=625
Now that we’ve taken a look at the kind of control that the Spectrum’s BASIC gives us over the hardware, it’s time to dip down into machine language and see what is offered to us there. There’s quite a bit more to cover in this realm, and I expect it’s going to end up spread out over quite a few posts. It is also quite a bit more fraught than on many other platforms, because Sinclair only ...

Replacing the Kobo Libra H2O Battery

https://ei3lh.eu/wp-content/uploads/2025/11/6b6b3f7e-9b96-4505-ab02-f2d7569cb651-1024x558.jpg
User had a Kobo Libra H2O with a faulty battery, which was difficult to replace due to lack of suitable options. They found a replacement battery on Ali Express and successfully replaced it with a soldering iron and basic tools.

How Hollywood stopped making movies in Hollywood

https://substackcdn.com/image/fetch/$s_!PU5R!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F4ed3afc1-5576-4677-aa1a-bf941416e9d7_1548x1024.webp
The entertainment industry in Los Angeles has been experiencing a production migration due to rising costs and tax incentives offered by other cities, leading to a decline in physical production in Hollywood. This migration has resulted in a widening gap between where stories are set and where they are filmed, with many productions choosing locations based on financial considerations rather ...

Kelly Criterion Simulator

Massively Parallel Postgres Backups

https://planetscale.com/assets/massively-parallel-postgres-backups-social-apuT3KE7.png
PlanetScale's backup system for sharded databases uses parallelism to minimize production impact and achieve petabyte-scale backups in hours, with techniques like spinning up backup-specific nodes and replaying Write-Ahead Logs. The system uses object storage, WAL archiving, and hybrid replay to ensure consistent backups, and is designed to be transparent and easy to use, with automated ...

AirLLM 70B inference with single 4GB GPU

https://raw.githubusercontent.com/lyogavin/airllm/main/assets/airllm_logo_sm.png
AirLLM is a library that reduces inference memory usage for large language models, allowing them to run on a single GPU card without quantization or pruning. It supports various models, including Llama, Qwen, DeepSeek, and more, and can be used with a simple one-line initialization.

Bonsai: Janestreet's UI Library

https://raw.githubusercontent.com/janestreet/bonsai/master/docs/assets/bonsai-logo.png
Bonsai is a UI library for building performant, reactive web applications in OCaml, inspired by Elm. It allows for composable state machines and incremental rendering, making it easy to manage state and UI updates.

ZX Spectrum System Tour: Sound

The ZX Spectrum has a 1-bit beeper and a General Instrument AY-3-8910 synthesizer chip for sound production, with the latter available in later models and as a peripheral for earlier ones. Three programs are created to play a C-major scale using the beeper and the AY-3-8910 chip, with the latter providing more precise and flexible sound control.

KisakCOD – open-source reimplementation of Call of Duty 4 Multiplayer

https://raw.githubusercontent.com/SwagSoftware/KisakCOD/master/GPLv3_Logo.png
Copy COD4 game files to bin/(BUILD_TYPE)/* for development. Use AI for assistance, but remain responsible for code committed.

The Billable Usage API: programmatic cost visibility for Cloudflare

https://blog.cloudflare.com/_image?href=https%3A%2F%2Fblog.cloudflare.com%2F_emdash%2Fapi%2Fmedia%2Ffile%2F01KZ1WZB6KJGKY7A4HBARJW87R.png&w=1999&h=1132&f=webp&fit=cover&position=center
Cloudflare launched a Billable Usage API for self-serve accounts, providing a single endpoint to return account usage and cost, broken down by product and service period. This API is designed for automation, allowing developers to access usage data in real-time, enabling FinOps teams to attribute costs to internal projects and teams.

Twenty Years of Pandoc

John MacFarlane created the document converter pandoc in 2006 using the Haskell programming language, which he chose for its strong type system and clean representation of structured documents. Over the next 20 years, pandoc evolved into a powerful tool supporting 51 input and 76 output formats, with over 600 contributors and a reputation for reliability and determinism, despite the potential ...

DDoS against Norwegian government IT infrastructure – status

Digitaliseringsdirektoratet's Status Page - ID-porten, Kontakt- og reservasjonsregisteret, Maskinporten, MinID, eFormidling, ELMA, eInnsyn, Ansattporten, Selvbetjeningsløsninger, Altinn - Driftsproblemer.

Prevent cognitive debt by manually retyping LLM-generated code

The user uses coding assistants to fast-forward through boring parts of projects, but manually types generated code to understand and adapt it, valuing comprehension over productivity. This workflow helps them build a mental model of their codebase and detect potential issues, allowing them to work faster and more efficiently.

Use Task Runners for Common Coding Tasks

https://hamvocke.com/img/default-preview.jpg
The user wants to create a convenience tool to run common tasks across multiple code repositories with different tech stacks, such as installing dependencies, building, linting, formatting, running tests, and deploying. They discuss various options, including bash scripts, make, just, and mise, to create a task runner that simplifies their workflow and allows muscle memory to be applied to ...

SearXNG in Rust

https://raw.githubusercontent.com/MikeLuu99/searxng-rust/main/assets/tui_screenshots.png
A Rust-based metadata search engine aggregates results from multiple search engines, deduplicates and ranks them using Reciprocal Rank Fusion. It supports concurrent querying, HTML scraping, and a ratatui-based TUI interface.

Rust project goals: Immobile types and guaranteed destructors

https://opengraph.githubassets.com/33d3cf50b3908ad51294f7ade72bf32057f24a694ef2052ca1c523d336cff23a/rust-lang/rust-project-goals
Rust proposes introducing new traits like Move and Forget to make explicit what operations are possible on a type, allowing types to opt out of being moved or forgotten. This change aims to simplify immovable types and enable safe scoped spawn for async, while eventually deprecating the Pin trait.