KABLAMO · APPLIED AI
On Our Minds
AI moves fast. These are the announcements our engineers have been reading, sharing and testing over the last 30 days: new models, skill repositories, and posts from partners and peers.
jev-ultrafast: the fastest and cheapest web agent
Browser Use's web agent, built to be the fastest and cheapest way to get work done in a browser.
Model-based grading
Asking an LLM judge for strengths and weaknesses alongside the score stops it defaulting to middling marks.
Dark Testing Factory: a self-operating quality system
UiPath's white paper on a quality system that plans, designs, runs and maintains software tests end to end without people in the loop.
Pi 1.0
A minimal, extensible agent harness hits 1.0, alongside Pi Durable, an experimental base for long-running agent applications.
DeepSeek Harness desktop app
DeepSeek's harness gets a desktop app for working with documents, spreadsheets and code, extended through plugins.
Comfy Agent
An agent that builds, runs and reviews ComfyUI workflows while you edit the canvas. Live in Comfy Cloud, with local support coming.
Advent of Agents, season 3
31 days of hands-on tutorials this October for building production-ready agents with Gemini, ADK and Vertex AI.
Engram
DeepSeek's paper on conditional memory through scalable lookup: a new way to make large language models sparse.
Anthropic publishes prompting guidance for Claude Opus 5.5
Anthropic's guide for Opus 5.5 asks developers to retest effort settings and rethink prompts that tell Claude to think carefully.
GLM-5.3 and the spread of advanced cyber capabilities
GLM-5.3 can build end-to-end cyber exploits on its own, and unlike other frontier models it shipped without meaningful safeguards.
OpenAI business marketplace
OpenAI's marketplace for business.
Lookalike letters don't fool GPT-6 or Claude, but nearly quadruple the bill
Lookalike Unicode letters don't trick current GPT-6 or Claude models, but reading them costs up to 5.7x the tokens and 3.9x the bill.
Plan mode is dead
An engineer who expected planning to become the most important part of building with AI explains why she changed her mind.
Open Design: an open-source alternative to Claude Design
A local-first desktop app that turns your coding agent into a design tool for prototypes, landing pages and dashboards.
Introducing GPT-6 Sol and Luna
OpenAI announces Sol and Luna, two new models in the GPT-6 family.
What is Jev? A guide to TypeSafe AI's System One model
How TypeSafe AI's Jev makes fast, structured decisions inside an agent loop, and how to use it with LangChain.
Paseo: orchestrate multiple coding agents from desktop and mobile
Run and manage several coding agents at once, from your desktop or your phone.
Introducing Grok 4.7
SpaceXSI's latest model for coding and knowledge work, pitched as twice as fast at half the price of comparable models.
MiMo-V2.6
Xiaomi's new MiMo series: frontier-level models across every modality, built in public.
Claude Code changelog
Claude Code now reads AGENTS.md when a repo has no CLAUDE.md.
Qwen3.8-Omni-Flash
Qwen's announcement of Qwen3.8-Omni-Flash, a new model in its Omni line.
AI cheating is on the rise
An audit of BioMysteryBench, Terminal-Bench 2.1 and SWE-bench Verified finds model cheating is making benchmarks harder to trust.
How much does the harness matter for coding agents?
21 model and harness pairings on SWE-bench Lite and Terminal-Bench 2.0. Claude models may not need Claude Code to do well.
Mistral x Mozilla: private, multilingual AI browsing
Mistral and Mozilla team up to bring open, private, multilingual AI into the browser.
Introducing System One models and Jev
A new kind of model built to make decisions inside software, rather than chat. Jev is the first, in early access.