Founder of Totally Werks, an independent software studio. We design, build, and operate internet products.
Minneapolis, Minnesota
Joined September 2026
- Tweets171
- Following133
- Followers54
- Likes205
Pinned Tweet
Founder of Totally Werks, a small independent software studio in Minneapolis. I design, build, and operate internet products.
Expect posts on what I'm building, what I learn running a studio alone, and the occasional opinion on tools.
Greg Porter retweeted
MotorStorm: Pacific Rift released on PS3 18 years ago today.
Racing through jungles, dodging lava, and sending monster trucks flying off cliffs was absolute chaos. Every vehicle had its strengths, and the different routes made every race unpredictable.
I miss this series a lot
Can't wait to get a proper office space set up when moving in a few months. Anyone have ideas or suggestions?
This could be extremely helpful for folks with large scale games. May have to spin up a new project now.
For all the TMBB folks, Artificial Analysis now grades coding agents on whether a maintainer would merge the change. Passing tests was always the low bar. Opus 5.5 in Claude Code leads at 61.7. It also costs $11.42 a task.
Announcing Artificial Analysis Coding Agent Index v2.0: FrontierCode v1.1 from Cognition replaces DeepSWE, SWE-Atlas-QnA grading is revised, and Claude Opus 5.5 in Claude Code leads at 61.7
Today we are launching v2.0 of the Coding Agent Index. The Index evaluates models together with their agent harnesses, now across 290 tasks covering software engineering, terminal use, and repository Q&A, with the three benchmarks weighted equally.
Changelog (Index v1.5 → Index v2.0):
→ FrontierCode v1.1 replaces DeepSWE v1.1. FrontierCode is a benchmark from @cognition with a private test set: 100 software engineering tasks on real open-source repositories, graded on whether a maintainer would merge the change. An attempt that fails a blocking check scores zero, and one that passes them all scores between 70% and 100% depending on the non-blocking checks it meets. We also scan each trajectory with a modified version of Cognition's rule-based scanner, and any attempt flagged for accessing solution material scores zero.
→ Revised SWE-Atlas-QnA grading, now SWE-Atlas-QnA-AA. Scale AI's 124 repository Q&A tasks are unchanged. We reworded 127 of the 1,306 rubric criteria for more consistent verdicts and moved to GPT-6 Luna (xhigh) as the judge, with five votes per criterion. The dataset and judge configuration are published on Hugging Face.
Key results:
➤ Claude Opus 5.5 leads the Index: Claude Code with Claude Opus 5.5 (max) scores 61.7, followed by Claude Code with Claude Sonnet 5.5 (max) at 59.3, Muse Code with Muse Wonder Full (max) at 58.9 and Claude Code with Claude Fable 5.1 (max with fallback) at 57.6.
➤ FrontierCode v1.1 is where Opus 5.5 pulls ahead: Opus 5.5 scores 56.0% on FrontierCode, ahead of Codex with DeepSeek V4.1 Flash (max) at 52.6%, Codex with GPT-6.1 Sol (max) at 49.4% and Codex with GPT-6 Astra (max) at 49.3%. Below the top, FrontierCode scores are tightly packed: 15 of the 24 configurations on the Index score between 43% and 49%.
➤ SWE-Atlas-QnA-AA grading is stricter and more consistent: across the 21 configurations evaluated on both Index versions, the mean score falls 1.7 points, from 58.8% to 57.1%, with 14 of 21 moving down. Individual configurations move by up to 7 points: Muse Code with Muse Spark Open (xhigh) falls 7.0, Kimi K3 in Kimi Code CLI and Codex with GPT-6 Luna (max) 6.2, and Muse Code with Muse Wonder (max) 5.6, while Codex with GPT-6.1 Sol (max) gains 2.1.
➤ Four configurations sit on the cost frontier: Codex with DeepSeek V4.1 Flash (max) scores 47.3 at $0.39 per task, Codex with GPT-6.1 Sol (max) 54.1 at $1.50, Codex with GPT-6 Astra (max) 54.7 at $7.04 and Claude Code with Claude Opus 5.5 (max) 61.7 at $11.42. Costs are pay-per-token API costs, not subscription prices.
We will keep updating the Coding Agent Index incrementally, with more evaluations and agents to come.
Neat but I don’t think I’d appreciate my doctor saying “My mistake. moving on.” when I still feel sick.
Greg Porter retweeted
DeepSeek raised what Anthropic loses monthly.
Gotta pump those numbers DeepRookie.
Greg Porter retweeted
Implications of AI, in relation to software security:
#1 CVE discovery sped up ~10-100x vs before
#2 Manual patching needs to die... but how?
#3 Security monitoring needs to improve massively
#4 Defenders will need to use a LOT more AI
#5 Security will cost a LOT more
And #4 is an interesting one, given that many security professionals heavily dislike AI (and X: they're on Mastodon!)
But the longer security teams live in denial about AI, the more their systems will be vulnerable for attackers to access, without them ever knowing. Like how TiDB was practically an open house for attackers to come-and-go for a whole month...
Waking up this morning just to rotate keys. Yay.
An urgent notice from the Railway team
blog.railway.com/p/security-…
