Introducing Claude Haiku 5.5: The Cheapest, Fastest, and Most Capable Small Claude Yet
Claude Haiku 5.5 is a major upgrade for high-volume AI workloads, but switching from Haiku 4.5 requires more than changing the model ID. Her...
Oct 7, 2026
![First Post](data:image/svg+xml;utf8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100"><defs><linearGradient id="gFirst" x1="0%" y1="0%" x2="100%" y2="100%"><stop offset="0%" stop-color="%238b5cf6" /><stop offset="100%" stop-color="%23ec4899" /></linearGradient><filter id="fFirst"><feDropShadow dx="0" dy="2" stdDeviation="3" flood-opacity="0.3"/></filter></defs><circle cx="50" cy="50" r="45" fill="url(%23gFirst)" filter="url(%23fFirst)"/><circle cx="50" cy="50" r="38" fill="none" stroke="rgba(255,255,255,0.4)" stroke-width="2"/><path d="M60 30 C 50 30, 40 40, 35 55 L 30 70 L 45 65 C 60 60, 70 50, 70 40 C 70 35, 65 30, 60 30 Z" fill="%23ffffff" opacity="0.9"/><path d="M35 55 L 45 65" stroke="%23ec4899" stroke-width="2"/><circle cx="65" cy="25" r="3" fill="%23fef08a"/><path d="M65 15 L65 20 M65 30 L65 35 M55 25 L60 25 M70 25 L75 25" stroke="%23fef08a" stroke-width="2" stroke-linecap="round"/></svg>)
![Word Warrior](data:image/svg+xml;utf8,<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100"><defs><linearGradient id="gWord" x1="0%" y1="0%" x2="100%" y2="100%"><stop offset="0%" stop-color="%23b91c1c" /><stop offset="100%" stop-color="%23f87171" /></linearGradient><filter id="fWord"><feDropShadow dx="0" dy="2" stdDeviation="3" flood-opacity="0.3"/></filter></defs><circle cx="50" cy="50" r="45" fill="url(%23gWord)" filter="url(%23fWord)"/><circle cx="50" cy="50" r="38" fill="none" stroke="rgba(255,255,255,0.4)" stroke-width="2"/><path d="M25 60 L75 60 L70 75 L30 75 Z" fill="%23ffffff" opacity="0.9"/><path d="M50 20 L55 30 L55 70 L45 70 L45 30 Z" fill="%23ffffff" filter="url(%23fWord)"/><rect x="35" y="65" width="30" height="5" fill="%23b91c1c"/><rect x="45" y="70" width="10" height="15" fill="%237f1d1d"/></svg>)
11 min
Subscribe to Anshu Pathak's Newsletter
Get new articles and engineering posts delivered to your inbox.
EmbeddingGemma 2: One Small Open Model for Searching Text, Code, Images, Video and Audio
EmbeddingGemma 2 is a 740M-parameter open model from Google that searches five kinds of content with a single set of vectors. Here's what's ...
Oct 7, 2026
8 min
CASL: One Set of Permission Rules for Your Whole JavaScript App
CASL lets you write permission rules once and use them in your UI, your API and your database queries. Here is how it works, where it fits, ...
Oct 5, 2026
9 min
An OpenAI safety lead quit. The part worth reading is about the agents.
David Robinson, who led the writing of OpenAI's launch safety reports, left the company and published an essay calling its culture broken. H...
Oct 4, 2026
5 min
Cloudflare wants you to build the next Git platform
Cloudflare put Artifacts, its versioned filesystem that speaks Git, into open beta and launched a build competition around it. Here is what'...
Oct 3, 2026
3 min
How to Choose a Database in 2026: Eight Options Compared by Use Case
No database is best at everything. This guide compares eight popular ones by use case, license, current versions and trade-offs, so you can ...
Oct 1, 2026
11 min
Gemini 4 Argon: Google's new flagship is here, and almost nobody can use it yet
Gemini 4 Argon is Google’s new frontier model, leading on knowledge work, long-context reasoning and several benchmarks. But coding is mixed...
Oct 1, 2026
7 min
AI Wrote the Code. Now Nobody Understands It.
Code that works and passes tests can still be code nobody understands. Here is what a 1985 essay, a controlled study and a 2026 interview st...
Sep 30, 2026
5 min
Claude Sonnet 5.5: Same Price, Up to 30% Less per Task
Anthropic released Claude Sonnet 5.5 on September 28 at the same price as Sonnet 5. It claims faster output, up to 30% lower cost per task, ...
Sep 28, 2026
7 min
Sentry in 2026: The Complete Guide to Errors, Tracing, Replay, Logs, Seer, Costs, and Self-Hosting
Learn how Sentry works in 2026 with hands-on Next.js and NestJS setup, distributed tracing, Session Replay, Logs, Seer, monitoring, pricing,...
Sep 28, 2026
27 min
Google Is Testing a "Buy" Button for Flipkart Inside Gemini, and It's Not the Checkout It Promised
Google is testing a Buy button that lets some Indian shoppers complete Flipkart purchases inside Gemini and AI Mode. The catch: it's a Flipk...
Sep 27, 2026
7 min
Introducing Meta VR Glasses: A New Era for Virtual Reality
At Meta Connect 2026, Zuckerberg unveiled Meta VR Glasses — a 100-gram glasses-style headset with a tethered compute puck, 5K micro-OLED dis...
Sep 24, 2026
8 min
Xiaomi Just Open-Sourced the Best-Ranked Open-Weight AI Model in the World. What's Inside It.
Xiaomi's MiMo-V2.6-Pro just topped Artificial Analysis's Intelligence Index among open-weight models, tied with Grok 4.7, at a fraction of t...
Sep 22, 2026
8 min
Building an AI That Never Stops Learning on 8GB of VRAM
Most language models are trained once and frozen forever. mini-AGI is a small byte-level model that keeps learning continuously on a single ...
Sep 21, 2026
6 min
Best Open-Weight LLMs, September 2026
Which open-weight LLM should you actually run in September 2026? This guide checks the leading models against primary sources — model cards,...
Sep 20, 2026
19 min
AWS says it can't restore some data from Middle East facilities struck by Iran
AWS confirmed it can't restore its Bahrain cloud region or a UAE availability zone damaged by Iran-linked strikes earlier this year. Six mon...
Sep 17, 2026
4 min
Gemini 3.8 Live Extended Thinking vs GPT-Live-1 vs Grok Voice Think Fast 2.0: A Developer's Buying Guide
Google, OpenAI, and xAI each released a flagship voice-agent model between late July and mid-September 2026. The headline benchmarks look cl...
Sep 16, 2026
10 min
$400 Billion Gone: Wall Street Spooked by AI's Own Warnings
After Anthropic's Dario Amodei and OpenAI's Sam Altman both called for a slower pace of AI development, Nvidia fell into correction territor...
Sep 15, 2026
4 min
How Claude Fable 5.1 Cracked a 370-Year-Old Cipher
For 373 years, a 64-number cryptogram at the end of a 1653 book resisted every codebreaker who tried it. This past August, Claude Fable 5.1 ...
Sep 14, 2026
4 min
Consistent Hashing: Why Adding One Server Shouldn't Move Everything
Add one server under naive hash%N and most keys move for no reason. This post builds a real consistent hashing ring, measures how much less ...
Sep 10, 2026
6 min
Bloom Filters: The Data Structure That's Allowed to Lie (A Little)
Chrome and Cassandra both use Bloom filters to cheaply rule out 'definitely not here' before an expensive lookup. This post builds one from ...
Sep 8, 2026
5 min
Inside Praxist: The Boundary Architecture Behind an Autonomous Research Agent
Praxist keeps its core strictly separate from task-specific plugins, then runs parallel peers through a Deep Innovation Gate and quality-div...
Sep 7, 2026
9 min
The JavaScript Event Loop: Why Your Code Doesn't Run the Way You Think
Ever wondered why a Promise resolves before a setTimeout(fn, 0) callback, even though the timeout was scheduled first? This post breaks down...
Aug 25, 2026
4 min