marchattonhere Profile Banner
Marc Hatton Profile
Marc Hatton

@marchattonhere

Followers
327
Following
4K
Media
71
Statuses
1K

Product hobbyist. Thinker, Trainer, Prompter, Sigh. Long suffering @arsenal fan. let's facts get in the way of a good story.

onchain
Joined August 2021
Don't wanna be here? Send us removal request.
@marchattonhere
Marc Hatton
2 months
my PM productivity suite: - @obsdmd as a wiki for my messy mind (.md files) - @chatprd dump all my rough notes get a banger PRD in no time - @magicpatterns for prototyping beautiful UIs and landing pages - @claudeai (Code) prototype to production, ingesting Obsidian files
7
10
141
@marchattonhere
Marc Hatton
3 days
I wonder when risk based will be rolled out. I should be able to just say fox all the linting errors without having to queue 100s of continues But also I don't wanna let loose on pervasive / risky changes I'm sure the OpenAI team are thinking deeply about this.
@gdb
Greg Brockman
3 days
Codex for fixing thousands of issues overnight:
1
0
1
@usemonologue
Monologue
3 days
Don't stay trapped in a bad relationship. Be like Marc. We’ll buy you out—get 50% off an annual Monologue subscription when you make the switch. Send proof of your current subscription → hello@monologue.to to claim.
@marchattonhere
Marc Hatton
4 days
I've split up with WisprFlow. It felt like every longer note would fail transription. And i am LOVING Monologue. Amazing work @naveennaidu_m and co.
3
4
28
@marchattonhere
Marc Hatton
3 days
I prefer 100 tabs in my browser
@frankcost_
Frank
4 days
people of notion — a question: if you use @NotionHQ in the browser what makes you choose it over desktop? seeking to understand reply in thread pls ❤️
0
0
1
@marchattonhere
Marc Hatton
3 days
Our brains are LLMs, what is an original thought even?
@ryancarson
Ryan Carson
3 days
People who say that an LLM doesn't produce original thoughts need to explain how they're not just a result of their own training data set and context window
0
0
1
@dennizzer_k
SPIKE⚡
4 days
🚨Raya is considering leaving Arsenal in the summer in search of playtime saying he keeps training hard but he has no job on the pitch
671
2K
29K
@marchattonhere
Marc Hatton
4 days
@naveennaidu_m small bug: yday I used Monologue but it doesn't show up on the GitHub style activity tracker. Note a biggie, just bringing to your attention
1
0
0
@marchattonhere
Marc Hatton
4 days
I've split up with WisprFlow. It felt like every longer note would fail transription. And i am LOVING Monologue. Amazing work @naveennaidu_m and co.
2
1
9
@calcsam
Sam Bhagwat
4 days
ts ai conf launch #2 we keep getting asked about patterns for using @aisdk ui, @assistantui, and @CopilotKit with mastra so we built UI Dojo, a repo + frontend with mastra agents implemented in each of these we are committed to first-class support for all of them
10
15
65
@marchattonhere
Marc Hatton
5 days
sometimes you just need a pen and paper
0
0
1
@marchattonhere
Marc Hatton
6 days
Same as photos from a music concert
@rrhoover
Ryan Hoover
6 days
I don’t think I’ve ever looked at my Twitter bookmarks. Just me?
0
0
0
@marchattonhere
Marc Hatton
6 days
first thought it was a phishing email - seemed to good to be true. going to make some very green squares on github these next few weeks
@adocomplete
Ado
6 days
If you're a Claude Pro or Max user and haven't tried out Claude Code on the web, there's no better time. Head over to https://t.co/50rnKrJ0zE and look in the right hand corner. We're giving Pro and Max users $250 and $1000 in free credits to push CC on the Web to the limit!
0
0
0
@marchattonhere
Marc Hatton
6 days
when AI is your job interviewer a groundbreaking study of 70,000 job interviews has shown that 78% of candidates prefer AI interviewers over humans. source @NPR
1
0
1
@addyosmani
Addy Osmani
9 days
New in @ChromeDevTools! Debug an app's full performance trace with Gemini! After recording a trace, you can now chat with Gemini about the entire trace, related Performance insights, and even connected field data - all without needing to select specific context beforehand!
89
362
3K
@marchattonhere
Marc Hatton
8 days
And ask if to encase SQL using "with as"
@peer_rich
Peer Richelsen — oss/acc
8 days
how to write complex SQL ask AI copy & paste error ask AI copy & paste error ask AI
0
0
0
@marchattonhere
Marc Hatton
8 days
vibe coding is the art of 1 PM creating the tech debt of 20 engineers
@DevLeaderCa
devleader
9 days
Before LLMs: Coding: 3 hours Debugging: 1 hour … .. . After LLMs: Coding: 3 minutes Debugging: 1 week
1
0
5
@marchattonhere
Marc Hatton
8 days
GPT-5-codex > Claude Opus 4.1
@natolambert
Nathan Lambert
8 days
Claude 4.1 Opus > Claude 4.5 Sonnet
0
0
0
@marchattonhere
Marc Hatton
8 days
after switching my workflows from claude code to codex, I used conductor less. Not anymore. Love, love, love how simple and effective their UX is for isolating tasks in different git worktrees. Their UX is so good that @cursor_ai copied large parts of it with Cursor 2.0
@charliebholtz
Charlie Holtz
9 days
Introducing Codex in Conductor!
0
0
0
@dennizzer_k
SPIKE⚡
9 days
🚨Arteta becomes the first manager to park the bus in opposition's half
@WhoScored
WhoScored
10 days
Arsenal have the most touches in the opponent's box (320) by a Premier League side this season. 🔴✨
65
2K
20K
@marchattonhere
Marc Hatton
10 days
loving the @github halloween vibes
0
0
0
@marchattonhere
Marc Hatton
11 days
context switching is a superpower that many PMs have. and underestimate their ability. in the world of AI agents, lean in. it's about to get even crazier
0
0
0