Claude Fable 5.2 and Mythos 5.2: Price, Speed and Access Guide
Claude Fable 5.2 is 20% cheaper than 5.1 and up to 30% faster. How it differs from Mythos 5.2, what fast mode costs and what to watch for when migrating.
Read
Blog
Notes on software, products and technology.
What we learn from our projects, the news of the industry and the tools we use, shared as we go.
65 postsFollow with RSS
Gemini 4 Carbon: Price, Performance and Access Status
The price, benchmark results and access status of Gemini 4 Carbon, Google's most capable model. A cost comparison with rivals and takeaways for businesses.
Read
LiteLLM Security Incident: What to Do About Version 1.107.0
LiteLLM version 1.107.0 was pushed to PyPI and Docker with malware. We explain how to tell whether you are affected and what you need to do.
Read
Claude Security Continuous Scanning: Security Updates as Your Code Changes
Claude Security now continuously monitors your repositories and rescans as code changes. How it works, what it costs and what to watch out for.
Read
Node.js Vulnerability: What to Do About 4 High-Severity Flaws
On October 21, 2026, Node.js fixed 10 vulnerabilities; 4 are high severity and one allows remote code execution. Here is which version you need to move to.
Read
GitLab Critical Vulnerability: What to Do About the CVSS 10.0 RCE
GitLab patched a CVSS 10.0 flaw that allows unauthenticated code execution. We explain which version you need to move to on self-managed servers.
Read
PHP Security Vulnerability: What to Do About the Two Critical Flaws OpenAI Found
We explain which version you need to move to for the two critical PHP flaws found through OpenAI's Patch the Planet initiative, and what your business should do.
Read
Indra OCR 1.1: Is a GPU-Free, 1-Million-Parameter OCR Worth It?
Indra OCR 1.1 runs without a GPU on a model of about 1 million parameters; we explain its speed, accuracy, language support and license terms with numbers.
Read
What Is Superintelligence? Definition, Timeline Forecasts and What It Means for Businesses
What exactly does superintelligence mean, when do experts expect it, and why is it used as a label so often? A short, clear guide for businesses.
Read
Python 3.15 Released: Lazy Import, UTF-8 and the New Profiler
Python 3.15 is out: lazy import, UTF-8 by default, a new profiler and a JIT improvement. When should you move to production?
Read
Startup support programs: A guide to free credits from big tech companies
Cloud, AI and GPU costs can be the biggest line in an early-stage budget. We compiled what AWS, Google, Microsoft, NVIDIA and Anthropic give startups, who can apply and how to avoid wasting the credits.
Read
Claude Startups Program: 1 Year of Free Claude Team and API Credits for Startups
Anthropic has opened its startup program to more founders. We walk through who can apply, what you get and how to apply, step by step.
Read
What is Claude Haiku 5.5? Price, performance and use cases
Anthropic's new small model Haiku 5.5 cuts the price to a tenth while making a big leap in computer use and knowledge work benchmarks. We compiled the numbers, its limits and what it's suited for.
Read
Claude API Credit: Free Monthly Balance for Max and Team Subscribers
Anthropic has added a monthly renewing API balance to Max and Team subscriptions. We explain who gets how much, how to link the credit, and why it doesn't apply in Claude Code.
Read
The Search Box That Can't Find a "Phone Cover" When You Sell "iPhone Cases": Semantic Search With Contrastive Embeddings
Your customer types "phone cover" and your site says "no results", because the product is called a "case". How embedding models learn meaning through contrastive training, and semantic search with PostgreSQL + pgvector, step by step.
Read
The Model That Thinks an Apple Is an iPod: Build Your Own Image Search With CLIP
CLIP is a contrastive model that lets you search photos with words, yet a handwritten note on an apple can fool it. How CLIP learns, how to search your photo archive in 30 lines of Python, and where it fails.
Read
Why Does AI Lie So Confidently? The Anatomy of Hallucination
Court rulings that don't exist, made-up sources, imaginary discounts promised to customers... What AI hallucination is, why it happens, what it has cost companies in real life and how businesses can protect themselves.
Read
Google Play's 12 Testers and 14 Days Rule: A Roadmap to Publishing Your App
Opened a new personal Google Play developer account and found the "Production" button locked? A step-by-step guide to the 12 testers and 14 days closed testing rule, applying for production access and avoiding rejection.
Read
Get Back Everything You Deleted in Git: A Recovery Guide With git reflog
Ran git reset --hard by mistake, deleted a branch, or had a rebase go wrong? Don't panic: Git almost never deletes anything right away. How to recover lost commits, branches and stashes with git reflog.
Read
Why Does Your Query Take 8 Seconds? Getting to 8 Milliseconds With a Database Index
Was your app fast at first, then slower as data grew? Usually the culprit is a missing index. How database indexes work, how to diagnose with EXPLAIN and the most common mistakes, with PostgreSQL examples.
Read
Is That Email Really From Your Boss? How to Spot and Stop Spoofed Email
Did the "urgent, pay this account today" email really come from your boss? How to spot fake emails in 30 seconds, and how SPF, DKIM and DMARC stop criminals from using your company's name.
Read
What Are Core Web Vitals? How Many Customers Is Your Slow Site Costing You?
Google scores your site's speed and experience with three metrics: LCP, INP and CLS. What they mean, what good values look like and how to improve your site, in plain language.
Read
What Happens When You Type a URL and Press Enter? A Web Page's 300-Millisecond Journey
The moment you type an address and press Enter, a chain begins that reaches servers around the world: DNS, TCP, TLS, HTTP and rendering. A step-by-step, plain-language look at how a web page loads.
Read
What Is Prompt Injection? The Hidden Instructions That Fool Your AI Agent
A single sentence hidden in white text on a web page can make your AI agent leak your data. What prompt injection is, how it works and how businesses using agents can protect themselves.
Read
Why Can't Computers Add 0.1 + 0.2? The Hidden World of Floating Point
Type 0.1 + 0.2 into your browser console: the result is 0.30000000000000004. It's not a bug; it's how computers store numbers. Why it happens, how it cost lives in a missile defense system, and how to protect your money.
Read
The Turkish I Problem: How One Dotted Letter Breaks Software Worldwide
Turkish's i-İ and ı-I distinction is one of the most famous sources of software bugs. Why does your app break on a Turkish user's computer, why does "title" become "TİTLE", and how do you protect yourself?
Read
How to Write an MCP Server: Connect Your Own Tool to Claude in 30 Lines of Python
Want Claude to reach your company's order system, database or internal API? A step-by-step guide to writing your own Model Context Protocol server and connecting it to Claude Desktop and Claude Code.
Read
9 Ways to Get Better Answers From Claude: A Prompting Guide
Same model, different prompt, completely different result. Based on Anthropic's own guidance: giving Claude context, XML tags, examples and room to think, with before/after examples.
Read
What Are Docker and Kubernetes? The End of "It Worked on My Machine"
Docker puts your app in a box that runs the same everywhere; Kubernetes manages hundreds of those boxes. The difference, how they work together and when you actually need Kubernetes, with examples.
Read
OpenAI Dots or Grok Bot? Comparing Always-On AI Agents
OpenAI's dots and xAI's Grok Bot implement the same idea in different ways: AI teammates with their own computers that work 24/7. Similarities, differences and who each one suits.
Read
How to Use OpenAI Dots: A Step-by-Step Guide From Setup to Custom Rules
You have access to OpenAI dots, now what? A practical guide to creating your first dot, connecting apps, scheduling tasks and setting custom rules.
Read
Has Your Password Been Leaked? How to Check Without Revealing It
"123456" has leaked 210 million times. How to check whether your password has been breached without sending it anywhere, and how websites should store passwords.
Read
What Makes a Strong Password? Will Yours Fall in 3 Seconds or 132 Years?
A random 8-character password holds for 132 years, a predictable one falls in 3 seconds. 2026 data, the new NIST rules and the formula for a truly strong password.
Read
Enterprise Software: Off-the-Shelf Package or Custom Build?
A package starts fast, custom software fits your work. Five questions to decide, plus how cost, security and data protection compare.
Read
Website or Web App? How to Choose for Your Business
A website informs, a web app gets work done. Five questions to find what you need, the key differences and an example from our own site.
Read
The Year 2038 Problem: Will Computer Clocks Stop in 12 Years?
On January 19, 2038, 32-bit Unix clocks will overflow and the date will jump back to 1901. We explain the Year 2038 problem, its GPS rehearsal and how to protect your systems today.
Read
Claude Message Batches API: Cut Your AI Bill in Half
The Claude Message Batches API processes requests in bulk and halves token costs. Here is how it works and when to use it.
Read
The Most Expensive Software Bugs in History: How One Line of Code Burned Billions
Bugs that blew up a rocket in 37 seconds, nearly sank a company in 45 minutes and crashed 8.5 million computers. No hackers involved; each was one small missed detail.
Read
How to Use Claude Code: A Guide From a Real Project
We build this website and its portal with Claude Code. Installation, CLAUDE.md, permission modes and what we learned from a real task.
Read
Mobile App Cost in 2026: How Flutter Changes Your Budget
We break down the items that determine mobile app cost and explain how Flutter affects the budget with example project scopes.
Read
GEO Optimization Guide: How Your Content Stands Out in AI Search
We explain step by step how GEO optimization makes your content visible and citable in AI-powered search engines.
Read
What Are SEO and GEO? A Complete Guide for Businesses
The answer to what SEO and GEO are determines your business's visibility both on Google and across AI assistants.
Read
Codex Security Cloud Update: What's New?
Codex Security Cloud gained continuous scanning and automated remediation features with the update announced at DevDay 2026.
Read
What Is OpenAI Dots? The New Agent Powered by GPT-6 Astra
OpenAI Dots is the new agent assistant powered by GPT-6 Astra. It can access more than 4,000 apps and complete tasks on your behalf.
Read
What Is GPT-6.1 Sol? Price and Performance Explained
GPT-6.1 Sol is OpenAI's new model, delivering performance close to GPT-6 Astra's level at one-fifth of the price.
Read
Gemini 4 Pro Leaks: Everything You Need to Know
Gemini 4 Pro leaks reveal the first traces of Google's new model, including token limits, benchmark results, and a possible release date.
Read
Claude Sonnet 5.5 Released! What You Need to Know
Claude Sonnet 5.5 is much faster and cheaper than Sonnet 5, while delivering performance close to Opus 5.5.
Read
What Is WebMCP? The New Web Standard for AI Agents
The answer to what WebMCP is: a new standard that makes your website accessible to AI agents. We explain how it works and how to set it up.
Read
Sonnet 5.5 Leak: What We Know About Anthropic's New Model
Sonnet 5.5 leak claims are spreading fast, but Anthropic still hasn't given an official date. Here's what's confirmed and what contradicts itself.
Read
EVREN, Türkiye's National Artificial Intelligence Platform
EVREN has launched as Türkiye's national artificial intelligence platform. We examined the details of its models, access, and credit system.
Read