SIRMON.AI
let understanding = Understanding::from(&release_notes) .pressure_tested_by(&commit_history) .verified_against(&source_code);
A personal site for agentic development, enterprise cloud
architecture, deep technical troubleshooting, and the occasional
hard-earned opinion.
Escalation Engineer: I work in the blast zone between customers
and engineering. I get the mess nobody fully understands...
incomplete evidence, customer pain, bugs loose in production.
...It's quite possibly one of the most satisfying jobs in tech
...until it's not
Goose
Bloodhound, morale officer, and unsanctioned network sniffer, pentester.
Meet Goose
Every serious operation needs one team member who ignores process, follows instinct, and still finds the weak point faster than everyone else.
People Worth Listening To
Smart people with useful things to say about AI, with a strong bias against nonsense.
This is the "less hype, more signal" section.
Scott Hanselman
The standard. Thoughtful, practical, and consistently worth the time.
People I Read To Stay Sharp
When I want signal over performance, these are the people I trust to explain what matters and why.
Still one of the clearest writers on how modern deep learning and LLM systems actually work. "Software 2.0" and "A Recipe for Training Neural Networks" are required reading.
One of the best writers on AI systems engineering, production ML, LLM architecture, and the workflow realities behind AI engineering.
Probably the best commentary blog on how people actually use AI, what it does to work, and where the productivity claims are real.
High Signal
Solid channels when you want practical AI and tech coverage instead of a weekly identity crisis about the future of software.
Personal favorite: the "let it cook" series.
Still Good, Minor Clickbait Penalty
Useful enough to keep in rotation. Every so often they catch me with a little clickbait, but the content is still solid.
Frontier Models
A practical snapshot of the model landscape I pay attention to: major closed frontier systems, serious open source contenders, and the emerging models that could matter next.
As of 2026, this is a working capability map, not a fan club.
Closed Model Leaders
The models still setting the pace for general capability, reasoning quality, multimodal performance, and production mindshare.
Open Weights Worth Respecting
The open models that are strong enough to matter strategically, not just as benchmark curiosities.
Watch Closely
The models and families that are not obvious incumbents yet, but could force a reshuffle with the next release cycle.
Maker Channels
The non-AI branch: the work I do, the craft I study, and the channels that keep the creative side sharp.
The Shark Artist
This one is mine. Chainsaw sculpture, art, and the hands-on side of the same brain that likes hard technical problems.
Channels I Follow For The Craft
The most directly relevant lane for me as a chainsaw sculptor: saw work, tree work, field experience, and people who have clearly paid their dues.
Hands-On Art Channels
The broader studio side: sculpture, fabrication, realism, painting, and visual design work that rewards patience and repeatable craft.
Sharks, Music, And Adjacent Rabbit Holes
The less directly related but still very on-brand group: shark world, music, and other corners of the internet that keep things interesting.
Storytelling Channels
Pure entertainment, strong pacing, and the kind of long-form story channels that can quietly eat a whole evening.
Find Me Online
The useful links, without the fake consulting landing page energy.