AI Intelligence Archive

Clinical Large Language Models (LLMs) achieve strong medical-exam accuracy; however, a correct answer does not guarantee that the explanation names the concepts that actually drove the decision. We introduce three lightweight, directly int…

Why adding autonomy to a software problem can make the architecture worse We have started calling...

Last week I got an unsolicited security report about my QR code service. It was polite, it was...

Background I have a LINE Bot that I use every day, linebot-helper-python. If you send it...

A deep dive into why I built EcoPrompt, the peer-reviewed research behind its metrics, and how to track LLM environmental impact locally in your browser.

I want to talk a bit about where we came from in software and where we are now. Not the industry in...

Evaluation is not a new discipline invented for a new kind of system. It is the oldest discipline in...

Most developers still treat prompt injection as a leakage problem. Someone types an adversarial...

Agent runtimes optimize for density and reject when full. That is not scheduling. Three failure modes at capacity, and who gets to call an agent...

LoRA and DoRA explained: matrix shapes, the geometry of weight decomposition, exact memory math for a 27B model, and what the DoRA benchmarks support.

Originally published on my personal...

This is a submission for the Kaggle Benchmarking Challenge What I Benchmarked I built...

The honest constraint behind everything I build is that there is one of me. A portfolio of sites...

I've been building Hexzie (HexTelegram) a Telegram-based AI assistant that goes beyond just sending...

A Claude Code Stop hook that runs your tests every time Claude tries to finish, and sends it back to work when they fail.

Most LLMs are built to generate. Jev is built to decide. That difference sounds small, but it...

Resumo em português: este tutorial mostra como dar a um agente de IA (Claude, Cursor, ChatGPT) acesso...

There's a special kind of stubbornness that leads someone to say "I will generate photorealistic 3D...

TL;DR The exposure problem: AI-driven development pushed exposed credentials to 1.27...

Every company adopting AI eventually runs into the same problem. An employee opens ChatGPT, Claude,...

Every time I asked an AI for an architecture diagram, I got the same thing back. A code block of...

TL;DR I changed one line in my coding agent's spec file and, two weeks later, it quietly...

Introduction Software is being built faster than ever. AI assistants can scaffold a...

A human in the loop is not the same as a human in the way. Here is the pattern for routing only the decisions that need a human, the four fields...

Retry LLM API calls on 429, 5xx, and timeouts with exponential backoff plus jitter, honor the Retry-After header when the provider sends one, cap...

I built TaskFlow, a project management app with an AI feature called Quick Add. You type "assign the...

This is a submission for the Kaggle Benchmarking Challenge. What I Benchmarked When an...

Most cost tracking keeps the attempt that worked and drops the ones that failed. Here is what that hides, and how to put every attempt on the bill.

A practical comparison of GitHub Copilot, Cursor, and Codeium based on coding workflows, context understanding, IDE integration, pricing, and real-world use cases.

Most classification and routing calls use a chat model, then parse prose and hope the JSON holds. The new /v1/systemone LLM decision API on LLM Gateway returns typed yes/no, choice, and score answers with calibrated probabilities, billed o…

A worked model of LLM cost optimization for a team spending $100k a month: what difficulty-based model routing saves at three traffic mixes, what the classifier costs, and the break-even point. Real catalogue prices, explicit assumptions, …

Automatic model selection now reads the request before it routes it. Choose the models `smart` may resolve to, and a classifier rates each prompt so a one-line fix goes to a cheap model and a hard problem goes to a frontier one. Available …

A live Smart Routing benchmark compares the same prompts across routed and fixed-model requests, including classifier charges, answer checks, and latency. See where routing saved money and where it missed.

Today's blog is mostly about the common ways cache misses have happened in our system, and how we...

The median across all 70 tested cases is 7.7x. BeautifulSlop is the compatibility API on top of a...

Codex and Aider look alike from the outside: you type, the model edits your code. Inside they are...

The experiment was fine. The runner was the bug. A single hung API call discarded hours of...

A benchmark on whether LLMs check WHO said something before acting on it. Strip the source label off a rumor, phrase it like policy, and a model that correctly refused will spend your money. #kagglechallenge

I review more code than I write now. I'm not sure I'm getting better at it. It's a small,...

Let me show exactly how it works with an example 🛻 The extension adds a new "ReptClip" tab in the VS...