#LLM
Curated coverage on «LLM»: releases, research, benchmarks, open-source.
Fintech group Svoi explains how to make LLM coding agents cheaper and more accurate
Fintech group Svoi has published a practical guide on turning an LLM from an "enhanced search" into a controllable agent, cutting token costs and improving code…
How a Habr author turned seven n8n workflows into an autonomous AI news system
In six weeks, the author turned a brittle set of seven n8n workflows into a single Python pipeline with 11 workers, 5 AI agents, a local LLM and Telegram control.
OpenAI engineer Sarang Gupta helps companies find customers and lift sales
OpenAI's Sarang Gupta builds marketing and sales models that show which channels bring in customers, where efficiency leaks, and how to speed up ChatGPT adoption at…
Yandex Code Assistant tested on a secrets-storage task and compared with Cursor
How Sovcombank Cut Its Product Team's Routine Work by 50% with an AI Assistant
Anthropic, Google and Microsoft quietly patched AI agent bugs with no CVEs or public warnings
US court rules a defendant's chats with Anthropic's Claude are not covered by attorney-client privilege
AWS shows how speculative decoding on Trainium2 speeds up generation in vLLM
AWS explains how speculative decoding on Trainium2 with vLLM can speed up LLM generation, cut per-token cost and raise throughput on long-output workloads.
Critics Call OpenAI's Partnership With McKinsey and Accenture a Bet on AI Hype
A scathing column argues OpenAI's Frontier Alliances sells businesses not a finished result but an expensive process of rolling out AI coworkers through big…
Google unveils Gemini 3.1 Flash TTS: a speech model with fine-grained control, dialogue and 70+ languages
Google releases standalone Gemini app for macOS, making AI easier to reach on Mac
Rufler simplifies swarms in Claude Code: one config instead of manual agent orchestration
Why Infosys Is Retraining Graduates: Indian Software Grads Aren't Ready for the AI Era
Selectel engineer shows an LLM agent that finds available domain names
A Selectel engineer built a Python service that asks an LLM to generate domain names and checks them via WHOIS right away, keeping only the available ones.
Anthropic explains when to start a new session in Claude Code, and how
Anthropic shipped a /usage command and explained how to manage Claude Code sessions so a million tokens of context doesn't turn into noise and degrade answers.