Tommi Somersuo
@tommis.fi
Director of Technology @ Solita.fi | Tech management & leadership, Dev, AI, agents, SWE, low-code, innovation and emerging tech.
Offline: Runner, angler, RES/SRA, Tibetan Terrier owner.
Offline: Runner, angler, RES/SRA, Tibetan Terrier owner.
Reposted by Tommi Somersuo
We're super jazzed to be launching GitHub Agentic Workflows into technical preview today!
"Generative AI" kind of fooled us into thinking that AI is synchronous, but async AI can have so much more value for us as developers. 🧵
github.blog/ai-and-ml/au...
"Generative AI" kind of fooled us into thinking that AI is synchronous, but async AI can have so much more value for us as developers. 🧵
github.blog/ai-and-ml/au...
Automate repository tasks with GitHub Agentic Workflows
Build automations using coding agents in GitHub Actions to handle triage, documentation, code quality, and more.
github.blog
February 13, 2026 at 8:46 PM
We're super jazzed to be launching GitHub Agentic Workflows into technical preview today!
"Generative AI" kind of fooled us into thinking that AI is synchronous, but async AI can have so much more value for us as developers. 🧵
github.blog/ai-and-ml/au...
"Generative AI" kind of fooled us into thinking that AI is synchronous, but async AI can have so much more value for us as developers. 🧵
github.blog/ai-and-ml/au...
Reposted by Tommi Somersuo
There are obviously limits to how much an agentic system can improve relative to the base model, but it’s really satisfying to nurture a stateful agent in its growth within those confines, and as a human you also discover new techniques from other humans:
claude.ai/public/artif...
claude.ai/public/artif...
Agent Memory Architecture 2026: Building Better AI Systems
Explore three memory types (episodic, semantic, procedural) that define AI agent capability. Learn consolidation strategies, retrieval optimization, and production architecture patterns for 2026.
claude.ai
February 14, 2026 at 3:49 PM
There are obviously limits to how much an agentic system can improve relative to the base model, but it’s really satisfying to nurture a stateful agent in its growth within those confines, and as a human you also discover new techniques from other humans:
claude.ai/public/artif...
claude.ai/public/artif...
Reposted by Tommi Somersuo
When working with Claude.ai skills there are ...
Numbers Every Claude.ai Container Skill Developer Should Know | Oskar 🕊️ | WhiteWind blog
Numbers Every Claude.ai Skill Developer Should Know
Inspired by Jeff Dean's "Numbers Everyone Should Know" — adapted for the Claude.ai Container Compute Environment. Empirically researched and written by Claude opus 4.6
Jeff Dean's insight was that engineers who internalize a few key performance n...
whtwnd.com
February 13, 2026 at 9:06 PM
When working with Claude.ai skills there are ...
Reposted by Tommi Somersuo
Very hard to find AI benchmarks that don't look like this
(and yes, that includes obscure benchmarks that nobody would train on and benchmarks with holdout datasets)
(and yes, that includes obscure benchmarks that nobody would train on and benchmarks with holdout datasets)
February 13, 2026 at 7:26 PM
Very hard to find AI benchmarks that don't look like this
(and yes, that includes obscure benchmarks that nobody would train on and benchmarks with holdout datasets)
(and yes, that includes obscure benchmarks that nobody would train on and benchmarks with holdout datasets)
Reposted by Tommi Somersuo
openai.com/index/harnes...
i have been mildly obsessed with the ideas in here for the past month or so.
incredibly validating, total victory: my suspicions are correct, actually. "This functions like garbage collection" is a sentence I have said out loud. craziness
i have been mildly obsessed with the ideas in here for the past month or so.
incredibly validating, total victory: my suspicions are correct, actually. "This functions like garbage collection" is a sentence I have said out loud. craziness
Harness engineering: leveraging Codex in an agent-first world
By Ryan Lopopolo, Member of the Technical Staff
openai.com
February 13, 2026 at 4:17 PM
openai.com/index/harnes...
i have been mildly obsessed with the ideas in here for the past month or so.
incredibly validating, total victory: my suspicions are correct, actually. "This functions like garbage collection" is a sentence I have said out loud. craziness
i have been mildly obsessed with the ideas in here for the past month or so.
incredibly validating, total victory: my suspicions are correct, actually. "This functions like garbage collection" is a sentence I have said out loud. craziness
resources.anthropic.com/hubfs/The-Co... "What you'll get out of this guide: By the end, you'll be able to build a functional
skill in a single sitting. Expect about 15-30 minutes to build and test your first working skill using the skill-creator"
skill in a single sitting. Expect about 15-30 minutes to build and test your first working skill using the skill-creator"
resources.anthropic.com
February 13, 2026 at 8:05 PM
resources.anthropic.com/hubfs/The-Co... "What you'll get out of this guide: By the end, you'll be able to build a functional
skill in a single sitting. Expect about 15-30 minutes to build and test your first working skill using the skill-creator"
skill in a single sitting. Expect about 15-30 minutes to build and test your first working skill using the skill-creator"
Reposted by Tommi Somersuo
It prevents wasted effort, keeps me in control of architecture decisions, and produces significantly better results with minimal token usage than jumping straight to code."
boristane.com/blog/how-i-u...
boristane.com/blog/how-i-u...
Boris Tane
The research-plan-implement workflow I use to build software with Claude Code, and why I never let it write code until I've approved a written plan.
boristane.com
February 12, 2026 at 7:13 AM
It prevents wasted effort, keeps me in control of architecture decisions, and produces significantly better results with minimal token usage than jumping straight to code."
boristane.com/blog/how-i-u...
boristane.com/blog/how-i-u...
Reposted by Tommi Somersuo
Time to consider not just human visitors, but to treat agents as first-class citizens. Cloudflare’s network now supports real-time content conversion to Markdown at the source using content negotiation headers.
https://cfl.re/4ksZQ1S
https://cfl.re/4ksZQ1S
Introducing Markdown for Agents
The way content is discovered online is shifting, from traditional search engines to AI agents that need structured data from a Web built for humans. It’s time to consider not just human visitors, but...
blog.cloudflare.com
February 12, 2026 at 2:32 PM
Time to consider not just human visitors, but to treat agents as first-class citizens. Cloudflare’s network now supports real-time content conversion to Markdown at the source using content negotiation headers.
https://cfl.re/4ksZQ1S
https://cfl.re/4ksZQ1S
Reposted by Tommi Somersuo
Anything you can do in Obsidian you can do from the command line.
Obsidian CLI is now available in 1.12 (early access).
Obsidian CLI is now available in 1.12 (early access).
February 10, 2026 at 3:14 PM
Anything you can do in Obsidian you can do from the command line.
Obsidian CLI is now available in 1.12 (early access).
Obsidian CLI is now available in 1.12 (early access).
Reposted by Tommi Somersuo
I wrote about the most ambitious form of AI-assisted software development I've seen yet - Strong DM's "Software Factory" approach, where two of the guiding principles are "Code must not be written by humans" and "Code must not be reviewed by humans" simonwillison.net/2026/Feb/7/s...
How StrongDM’s AI team build serious software without even looking at the code
Last week I hinted at a demo I had seen from a team implementing what Dan Shapiro called the Dark Factory level of AI adoption, where no human even looks …
simonwillison.net
February 7, 2026 at 3:42 PM
I wrote about the most ambitious form of AI-assisted software development I've seen yet - Strong DM's "Software Factory" approach, where two of the guiding principles are "Code must not be written by humans" and "Code must not be reviewed by humans" simonwillison.net/2026/Feb/7/s...
Reposted by Tommi Somersuo
what you want is an MCP. oh wait have you tried installing a skill? no you don't have an agents,md file. actually what you want is a plugin. have you tried agent orchestration? you have gotta try hooks.
February 6, 2026 at 4:54 AM
what you want is an MCP. oh wait have you tried installing a skill? no you don't have an agents,md file. actually what you want is a plugin. have you tried agent orchestration? you have gotta try hooks.
”limited-time $50 (USD, or local currency equivalent) in extra usage to Pro and Max users to coincide with the launch of Claude Opus 4.6.”
Claude Opus 4.6 extra usage promo | Claude Help Center
support.claude.com
February 6, 2026 at 4:30 AM
”limited-time $50 (USD, or local currency equivalent) in extra usage to Pro and Max users to coincide with the launch of Claude Opus 4.6.”
Reposted by Tommi Somersuo
🚀 VS Code is becoming your home for multi-agent development. A single place to view and manage all your agents - local, background and cloud, and even Claude or Codex, all under your Copilot subscription. Learn more ➡️ aka.ms/VSCode/109blog
Your Home for Multi-Agent Development
VS Code has become the unified interface for all your coding agents. Manage local, background, and cloud agents in one place, use Claude and Codex agents alongside Copilot, and benefit from open standards...
aka.ms
February 6, 2026 at 2:30 AM
🚀 VS Code is becoming your home for multi-agent development. A single place to view and manage all your agents - local, background and cloud, and even Claude or Codex, all under your Copilot subscription. Learn more ➡️ aka.ms/VSCode/109blog
Reposted by Tommi Somersuo
is it just me or is it getting hot in here?
Introducing GPT-5.3-Codex
GPT-5.3-Codex is a Codex-native agent that pairs frontier coding performance with general reasoning to support long-horizon, real-world technical work.
openai.com
February 5, 2026 at 6:33 PM
is it just me or is it getting hot in here?
