Dostoyevsky isn’t difficult — The Autodidacts

[autodidacts.io]Jun 26, 2026

Other than the names. The names are difficult.

A Practical Guide to SSH Tunnels: Local and Remote Port Forwarding

[labs.iximiuz.com]Jun 25, 2026

SSH port forwarding explained in a clean and visual way. How to use local and remote port forwarding. What sshd settings may need to be adjusted. How to memorize the right flags.

Building Reliable Agentic AI Systems

[martinfowler.com]Jun 24, 2026

AI helping pharmaceutical researchers query decades of information buried in PDF reports

una.im | Modern CSS theming with light-dark(), contrast-color(), and style queries

[una.im]Jun 24, 2026

Combine three new CSS features to build fully adaptive themed components.

In praise of memcached

[jchri.st]Jun 23, 2026

never waste a token

[sunilpai.dev]Jun 23, 2026

durable inference: resumable streams, crash recovery, and why the LLM request shouldn't die with your process.

The State of AI Post-Training Agents — Thoughtful

[thoughtfullab.com]Jun 23, 2026

An update to our FrogsGame report. With Claude Fable 5, Opus 4.8, and GPT-5.5, post-training agents improved sharply — Fable 5 fixed the biggest failure mode, low-quality SFT traces, by generating correct traces programmatically.

Everything Is Logarithms

[alexkritchevsky.com]Jun 23, 2026

Munich 1991: the Roots of the Current AI Boom

[people.idsia.ch]Jun 22, 2026

1991: T and P of ChatGPT, distillation, deep residual learning, LSTM, GAN

Fine Tuning a Local LLM to Categorize Questions

[teachmecoolstuff.com]Jun 22, 2026

pAs a fun personal project, I have been working on a chatbot for answering general questions about my household on anything from maintenance questions to doctor’s appointments. /p p The general idea is that the chatbot will get its household knowledge through RAG from querying a vector database, but for better results I have made the vector searches metadata aware. /p p Basically, I am running questions through a pre-processing step to categorize questions into known metadata categories (e.g. pool, car, hvac, cooking). The main goal of this is to narrow down the search space for vector ranking to only indexed entries that match the category of the question. As an example, the question “When did we replace our pool pump?” will be mapped to a category called “pool” before querying the Index database. /p

Load more