Builder and full-stack engineer. Built: 📷 https://xitter.cf/t.co/iFnkvKGdnm
Sydney, Australia
Joined February 2016
- Tweets29
- Following200
- Followers32
- Likes11.7K
Tables are easy to screenshot and annoying to use.
I wanted a faster way to make one editable without rebuilding every row by hand.
I tried it on a clean table and a deliberately rough scan. Here’s the workflow and exact Markdown it produced.
AI agents produce better work when you give them better context.
Snapdown captures visible text from documentation, PDFs, error messages, and tables, preserving headings, lists, code, and table structure as Markdown.
Paste it into Claude or Cursor instead of manually transcribing and reformatting it. One shortcut, entirely local.
A 25-row table trapped inside a PDF.
One shortcut later, it’s clean Markdown in Obsidian, with the rows, columns, and labels intact.
Snapdown is now on Macapp Supply.
See Snapdown on Macapp Supply:
macapp.supply/app/snapdown
I want to keep stress-testing it. What difficult table should I try next?
Today I launched Snapdown, a Mac app that turns anything on your screen into clean Markdown.
Unlike basic OCR, it keeps headings, lists, tables, formulas, and reading order.
One shortcut. Entirely local. Works offline.
10 captures free. $12 once.
snapdown.com.au/?utm_source=…
Broden Suffern retweeted
Stop renting your AI agent.
I'm a solo dev — I just open-sourced the kit to build (and own) your own. Any model, your machine, from a folder of text.
Watch it go from text file to live in 30 seconds 👇
#buildinpublic
Broden Suffern retweeted
Everyone's racing to build the AI agent — the coding one, the sales one, the support one. Each welded to a single job.
I've been building the opposite: one engine that becomes whatever agent the person in front of it actually needs.
Broden Suffern retweeted
Ask an AI agent to "understand this codebase."
It opens 4 files, runs 2 greps, says it's done.
That's not understanding. That's skimming and hoping.
Broden Suffern retweeted
Wallets on us, fr🙋♂️
Find us at @token2049, all attendees get a free SafePal X1 hardware wallet each🙌
📍 Booth PB4-8, Level 4
✅ Check in → complete simple tasks →walk away with an open-source Bluetooth hardware wallet
🚨While stocks last
🫡 See you soon
Broden Suffern retweeted
“The only real test of intelligence is if you get what you want out of life.” — @naval
Broden Suffern retweeted
So one thought about o3: the cost of running the model is extremely high (~$20 per task).
The assumption made by the Singularitarian cult predecessors of the Doomers, and by the Doomers themselves, was that there would be a vast "hardware overhang" (meaning that there would be a lot of spare hardware capacity out there when AGI was first developed), and that once you got AGI at all, it would very rapidly self improve because of all the excess hardware available that it could make use of.
I think at this point, we can finally dismiss that as fantasy; there isn't vast hardware overhang, we're in fact incredibly hardware limited these days, and can barely afford the hardware for cutting edge AI models.
Will the models become much cheaper to run? Over time, absolutely they will. Will we see some sort of fantasy hard takeoff happen in a situation where we are so hardware limited? No.
Broden Suffern retweeted
That just happened! Github Copilot FREE 🤯
2025 is shaping up to be the year of builders.
What are you going to build? 🚀
Broden Suffern retweeted
Your SaaS product without AI is like a T-Rex with a keyboard—technically functional, historically interesting, and completely extinct 🦖⌨️☄️
Broden Suffern retweeted
Plot twist: 'Devin Manager' might soon replace 'Dev Manager' in job listings.
Managing AI dev agents is about to become a whole new career path. Wild times ahead! 🤯
Broden Suffern retweeted
"We've scaled Ethereum 100x in the last year, with a path to 1000x. Growth requires temporary tradeoffs - just like Uber subsidized rides"
Fascinating discussion with @jessepollak on L2s, user adoption & the future of ETH scaling.
📺 Full interview: youtu.be/j5olRm0qRc8