SWE | Consultant | Entrepreneur
Joined September 2022
- Tweets 239
- Following 278
- Followers 296
- Likes 9,143
7 Photos and videos
12 Aug 2025
The way @AmpCode requires holding the command button when sending requests makes you so much more purposeful about what you pass to context...
But when you're coding hardcore while feeding a newborn baby, that becomes a bit of an issue š
3
16
5,064
23 Jul 2025
Ok, I donāt think Iāll ever go back to making hand-crafted, bespoke code. @AmpCode and claude code are too good
1
17
663
27 Jun 2025
SERIOUSLY, DO NOT BLINK
Now is the most important time to be staying at the cutting edge with SDLC using @claude_code and @AmpCode
Iāve had multiple 2Xs of productivity - just this year
Can finally focus on market and marketing - you know - what actually brings in the cash š
1
13
406
25 Jun 2025
With google, all knowledge is _available_. But it requires absorbing that knowledge, then mapping it onto your task to get the benefit.
With LLMs, the knowledge is also available - BUT - it also has cognition. So, you can skip the absorption step and direct it to your task
2
73
25 Jun 2025
Tried out SuperClaude today for a really tough UI issue I was dealing with. It was super easy to use, tell it where some screenshots of inspiration were and let the tokens flow
1
1
572
25 Jun 2025
81
20 Jun 2025
Getting this kind of error. Also, the agent is going off in crazy directions that don't make sense. Wasn't doing this yesterday, may've been a regression
33
Max Dignan retweeted
17 Jun 2025
here sitting in a meeting, a quick cheeky session here to X to post this
my laptop is sitting next to me and itās running two agents cooking on something
meetings are no longer expensive for software engineers
manager time/makers time is invalidated.
thatās the tweet.
thatās the ponderoo.
3
2
18
2,630
13 Jun 2025
Been researching @simonw's "Lethal Trifecta" and how it impacts coding agents like @cursor_ai @AmpCode or @claude_code
Thankfully, each blocked my *simulated* attacks. However, there may be some more improvements that could reduce user-approval-fatigue
maxdignan.com/writings/the-lā¦
3
9
1,168
13 Jun 2025
The biggest issue was that the best blocks for the lethal trifecta involves users blocking the request to exfiltrate info. Many users are running these agents in "Allow All" mode, and/or just clicking "Approve" without fully realizing risk
75
13 Jun 2025
Real excited to see the progress in Time Series Foundational Models
Was doing some research for this late last year using LLMs
Happy to see more focused models for this specific need parseable.com/blog/zero-shotā¦
54
29 May 2025
Ok, tried @AmpCode a couple weeks back.
Ran into some snags.
Retried it today, itās killer
Hands down the best agentic coder atm
2
17
2,144
Max Dignan retweeted
24 May 2025
Excited to see my LLM CLI tool used by Sean Heelan to help identify a remote zeroday vulnerability in the Linux kernel!
simonwillison.net/2025/May/2ā¦
13
34
569
54,989
12 May 2025
ādonāt use cursor, it might make you worse at codingā is the new ādonāt use google, you might not have it, so you should memorize.ā These tools arenāt going anywhere. Itās the new way
1
2
79
12 May 2025
Socrates had a similar complaint against a technology we all use everyday: the written word
1
32
28 Apr 2025
Stop calling it "vibe coding" like it's a bad thing. For engineers with hand injuries and attention challenges, it's the difference between having a career and not having one at all.
1
65
28 Apr 2025
This is truly freeing for engineers with permanent or temporary disabilities! Vibe coding isn't bad - it's simply another tool for engineers to utilize!
1
32
28 Apr 2025
I'm crafting more content for vibe coding skeptics. RT if you know someone who should see this! #VibeCoding #Accessibility #TechInclusion
30