@codergriff

I'm a Griff who likes to code. Having a blast using AI tools and agentic workflows for software engineering. This industry has never been so much fun.

 Aberdeen, Scotland
Joined September 2017
  • Tweets125
  • Following370
  • Followers140
7Media
Mark Griffiths retweeted
Nikhil Krishnan
@nikillinitOct 22
We’re like one hack away from me finally being able to get my medical records easily
69
296
27
5,773
141,873
Mark Griffiths retweeted
Kit Langton
@kitlangtonOct 4
I'm now having 𝚘𝚙𝚎𝚗𝚌𝚘𝚍𝚎 make a motion graphic explainer for every non-trivial PR.
80
38
18
1,439
92,560
Replying to @James_M_South
Could it be that you are doing some pretty rare software engineering not well represented in the training data? It’s not like you’re doing forms over data boilerplate
1
1
52
And they forget that humans produce slop too. And they do it often. Human sloppy code that’s full of lumps, bad smells, insects and the occasional stray hair
DHH
@dhhSep 28
"Slop" has become a comfort blanket for a cohort of programmers stuck somewhere between anger and bargaining on their way to acceptance. There might be some kicking and screaming, maybe a little crying, but eventually you have to let the blanket go and update your priors.
23
This one is amazing too
josh
@eudaemoneaSep 23
when Anthropic released their Functional Emotions paper, I gave it to Claude and asked for a song. tonight I asked Opus 5.5 to create a video for it. and it's breathtaking.
1
36
Do not nerf 🔥
anabology
@anabologySep 25
Gave Opus 5.5 donald's prompt, Midjourney, and a moodboard 12 hours later, woke up to this:
20
Mark Griffiths retweeted
eric provencher
@pvncherSep 15
I hate to say it, but if you’re running more than 2 sub agents at time, you’re almost certainly burning tokens for 0 quality gain. Agents don’t trust each other enough to avoid double-checking everyone’s homework.
250
49
59
1,290
286,294
The media: AI is uncontrollable and will kill us all Claude: The permission check blocked the tag push as a "Production Deploy". I've stopped rather than try to get around it
40
🤭
Jessie Frazelle
@jessfrazAug 7
All the AI labs claiming their models are hacking external systems. NOWS YOUR CHANCE TO PROVE IT HACK GITHUB AND FIX IT. The world thanks you.
56
Mark Griffiths retweeted
Tyler Denk 🐝
@denk_tweetsAug 6
owning a business is saying “after this week things will calm down” every week for the rest of your life
66
45
4
497
19,150
Mark Griffiths retweeted
amanda.eth
@amandacassattJun 5
When they zig, you should zag. That's why I recommend Thinking. Everyone's using AI so it'll really set you apart.
8
2
1
46
2,505
Manual dependency management incoming
Mark Griffiths
@codergriff11 Feb 2021
Superb write-up! Now can we please just go back to manually maintaining dependencies...
42
Pretty freaked out by this tbh
Simon Willison
@simonwApr 24
Replying to @simonw
Important: it has been confirmed that ChatGPT Images 2.0 added the "Why are you like this" sign of its own accord xitter.cf/scottjla/status/204754…
41
Replying to @amandacassatt
10/10 VG
1
24
This made me proper chuckle. A real genx relate lol
Eleanor Berger
@intellectronicaApr 2
i'd love to read the @AnthropicAI research on earworms. what layers light up in claude when it can't stop hearing the bangles' eternal flame playing in a loop in its mind?
1
53
Mark Griffiths retweeted
Corey Coto
@coreycotoMar 24
Enshrine your zone 💀
3
20
4,687
Mark Griffiths retweeted
Awni Hannun
@awnihannunMar 3
Claude Code (OpenCode / Codex / etc) is just so fun. It's really removed so much of the drudgery of writing software and left mostly the fun part. Almost makes me wish we could freeze capability development and enjoy this moment for a little longer.
17
16
3
322
14,857
Mark Griffiths retweeted
Jason Bosco
@jasonboscoFeb 27
"We used to debate using tabs vs spaces in code we'd type out"
108
1,036
82
11,923
378,436
Mark Griffiths retweeted
Ethan Mollick
@emollick24 Dec 2025
In fact the average ChatGPT query takes almost exactly as much energy as a Google search in 2008 (that is the last time Google clearly indicated the electrical consumption of a search).
169
186
28
4,500
261,764
Mark Griffiths retweeted
Mark Kretschmann
@mark_k21 Dec 2025
It's over. GPT-5.2 aces one of the most important benchmarks and it's not even close!
159
92
52
2,322
149,157
Load more