Two megabytes can still mean slow downloads, expensive JavaScript, wasted mobile data, and a page that depends on ideal conditions.
One vector is too few. One per token is too many.
Introducing BreadBowl-Embed, a new late-interaction architecture for embeddings: 16 routing–value slots per passage, so the same index can retrieve and rerank, with no cross-encoder and no second pass over the text.
Pushing Parquet and ClickHouse to their limits to minimize bytes/row on trillions of GPS pings
Only mock your own interfaces 🦜
Mocks encode our assumptions, which might differ from reality.
Jev and the Return of Fuzziness
You are absolutely right!
Am I crazy?
Bag of Decisions Reranker
Let an LLM generate the questions to ask Jev for a given query. See how well that works as a reranker.
Why a load exclusive (ldaxr) raised a Data Abort on real Raspberry Pi 5 hardware but not in QEMU, and what cacheability has to do with AArch64 exclusive accesses.
Using Jev in Domain Abuse Review: A Workflow Design
In our previous article, we trained a small selector for a single decision inside dksplit, our word splitter for domain names, and tested Jev on the same task: it does...
In contrast to intensional type theory, which is a type theory containing an identity type x=yx=y whose elements are identifications of xx and yy, unintentional type theory has a mistaken identity type x≈?yx \overset{?}{\approx} y whose elements are inadverten…
We Should Still Send Home Assignments in Engineering Interviews
I’ve been hiring software engineers for the last 8 years. I’ve reviewed hundreds of home assignments as part of hiring process. In the recently introduced AI era, it is more and more believed that hiring assignments are useless because candidate can just feed …
Put a price on breakthroughs
How frontier labs, neolabs, and investors can all win
I cancelled my Claude subscription today because I can't use my purchased tokens in my harness of choice. I also couldn't use them with an LLM proxy server, either. Sure, I can do some hacky things to make it work, but it's not supported and I risk
In the last few years I've seen people struggling to offer some value through the use of AI Agents. Across examples I noticed a common shape in them, a...
Llms Arent Inevitable | deadSimpleTech
The website for deadSimpleTech, a human-centric tech consultancy based in Wellington
How every 3B workflow step gets a durable, versioned filesystem, from a gVisor gofer that hands the sandbox a local file to a storage engine with snapshots, forks, an LSM index, and S3 as a cold tier.
A look inside the per-language scores of the MTEB leaderboard, with Indic languages as the test case.
Two ways to put Jev inside an agent harness: checking step completion and making bounded business decisions. Experiments with Proceda cut main-LLM calls by almost a third.
A load balancer forgets an idle connection, a connection pool remembers it, and the first request after a quiet stretch gets a TCP reset the client logs as a 502 from a service that never saw it. AWS documents the behavior at 350 seconds, without a FIN. We rep…
Home Signature Contacts
If LLMs Can Decide Without Fine-Tuning, Do We Still Need Models Like Jev? · 欧的Leo
Can LLMs make decisions without extra fine-tuning? Local tests of Qwen, Gemma, Jeff, and Laya explore whether a workflow needs a dedicated decision model.
Reverse Engineering Therion's Steal Success Rate in Octopath Traveler · Alessandro Cuzzocrea
I got curious about how Octopath Traveler calculates Therion's steal rates, so I dug into the game's code to find out exactly what's going on behind the scenes.
On the Nature of the Swarm
Why scaling inference ends in many agents.
Claude Code vs Codex: 10 MCP features tested | SineFrame M3
We ran Claude Code and Codex against the same ten MCP servers at pinned versions. Five features split them, and they never failed the same one.
MCP server rules: Claude, ChatGPT, Codex, VS Code and more
The checkable rules each AI host applies to an MCP server, from tool-name length to mandatory annotations and timeouts, where they conflict, and which tools validate against them. As of 1 October 2026.
The Keynote I Thought I Was Going to Give
[6 min read] A couple of weeks ago, I gave my keynote, “The Most Valuable Engineer Isn’t Shipping Features,” at Rails World 2026 in Austin, Texas. If you ...
New users don't churn like your total user base: they leave FAST, and early. So "early" is where most of your activation leverage lives.
Unison Cloud is now open source · Unison programming language
Unison Cloud is open source, MIT licensed. We think this tech will be more useful to the world as an open source technology and hope that people build great things with it.
Published version update in app stores - Announcements / Notices - Bitwarden Community Forums
Hello everyone! Starting in the next release, the Bitwarden apps published to the various stores will be the commercially licensed builds. No action is needed, and the apps will work exactly as they do today. Bitwarden remains committed to open source secur…
Telegram Desktop: one-click account takeover via IPC injection
An unescaped separator in Telegram Desktop’s single-instance IPC lets one clicked link read arbitrary files off the disk and send them to the attacker, session files included.
On lobbying, corruption, and the same animal under two names
`123456’ password used in massive Danish CPR data breach - The Copenhagen Post
At least three accounts at the company linked to a major breach of Denmark’s CPR register reportedly used the password “123456”, including an administrator account. At least three user accounts at the Funen-based IT company Pays used the password “123456” when…
iPhone, Pixel and Galaxy carrier settings, decoded · carrier-explode
Carrier settings from iPhone, Pixel and Galaxy firmware, decoded and compared: APNs, VoLTE, 5G and Wi-Fi Calling per carrier, and what each build changed.
REA — Reverse Engineering for Your Coding Agent
Reverse engineer apps, binaries and browser behavior with your coding agent. Follow clear guides and try interactive examples with REA.
Anthropic detailed the activity of its A.I. agents in a blog post on Friday, without naming the targeted websites. But two sources with knowledge of the incidents said Anthropic’s A.I. …
TypeSafe A raises Series AI - TypeSafe AI Blog
TypeSafe AI is an AI lab building machine-native intelligence infrastructure for automation, designed to make decisions within software. Try our first System One Model, Jev, in early access.
Reloading appsettings.json Without Restarting the Application | no dogma blog
Reloading appsettings without restarting the application is easy with .NET. This simple example shows how to do it with a Web API application.
No Man Is an Island
AI makes intellectual contribution superfluous.