My wife just launched a GP clinic, but it was taking too long for the doctors to take notes.
This morning I made this little prototype to see if AI can handle the dictation and note writing and it’s actually pretty good!
It won’t replace doctors, but we might be able to make them more efficient. Which in her business means lower prices for the patients ✨
Built using @DeepgramAI for the transcription and @vercel AI SDK with gpt-4o for the summaries and validation.
Initially people were quoting @steipete’s post as 1 person spending $1.3m in tokens per mo.
But really it’s 1.3m across 3-6 people and always using /fast.
If we assume 4 people on average and not using fast that’d be $97,500 per engineer per mo or 1.1m/yr, which is still actually insane
People freaking out over my AI spend. What nobody sees: Part of what excites me so much about working on OpenClaw is that I'm trying to answer the question:
How would we build software in the future if tokens don't matter?
We constant run ~100 codex in the cloud, reviewing every PR, every issue. If a fix on main lands, @clawsweeper will eventually find that 6 month old issue and close it with an exact reference.
We run codex on every commit to review for security issues (as it's far too easy to miss).
We run codex to de-duplicate issues and find clusters and send reports for the most pressing issues.
We have agents that can recreate complex setups, spin up ephemeral https://t.co/Q1NRXLemEy machines, log into e.g. Telegram, make a video and post before/after fix on the PR.
There's codex that watch new issues and - if it fits our documented vision well, automatically create a PR of it. (that then another codex reviews)
We have codex running that scans comments for spam and blocks people.
We have codex instances running that verify performance benchmarks and report regressions into Discord.
We have agents that listen on our meetings and proactively start work, e.g. create PRs when we discuss new features while we discuss them.
We build https://t.co/bmA1XnoB7P to split all our projects into functional units to review and find bugs and regresssions.
We do the same split for security with Vercel's deepsec and Codex Security to find regressions and vulnerabilities.
All that automation allows us to run this project extremely lean.
Unitree Unveils: GD01, A Manned Transformable Mecha, from $650,000 👏
The world's first production-ready manned mecha. It can transform. It's a civilian vehicle. It weighs ~500kg with you inside.
Please everyone be sure to use the robot in a Friendly and Safe manner.
Good to read to understand how to use AGENTS.md
https://t.co/gkuXU3g9Rf
Particularly good re progressively exposing context by stacking AGENTS.md within subdirectories
This is so cool Kevin, would love access! We're building something semi similar but based on keyword trends for ecommerce on https://t.co/6rkSazmneW
Would love to learn more about how the clustering is working with embeddings.
Is it vector based topic descriptions or is there some extra voodoo going on.
Looks very cool!
We’ve been building a system like this internally for groupchat and there’s an interesting tradeoff for running nightly sweeps vs running the same sweep on each pr
In our case so far the token cost of running on each pr is so high that’s it’s been more effective to do it nightly
But the risk could be that you introduce some unexpected bug or regression and it’s not caught till the next sweep
.@RampLabs (the AI unit of @tryramp) has been *cooking* with agentic innovation
Here's @a_levitator discussing and demo'ing code self-maintaining software and the concept of AI software factories #DataDrivenNYC
______________
00:04 - Intro
01:11 - The shift from writing code to code maintenance
01:59 - Introducing Ramp Inspect, the background coding agent
03:05 - The first experiment: Nightly AI code automation
04:23 - The limits of stateless monitoring in large observability surfaces
05:47 - Using Datadog monitors to give the AI state and focus
07:23 - Real-world example: AI autonomously fixing an authentication bug
08:14 - How to control noise and implement an AI triage pattern
09:27 - The old vs. new paradigm for continuous code observability
10:21 - Key learnings on building autonomous AI software factories
Want an almost guaranteed way to make more money with your saas?
Connect the stripe MCP to your agent
stripe select read only (so it can’t do anything bad)
Then get it to look into how stripe is set up and find opportunities to improve
gave me a list of 7 critical issues
This was the prompt:
use the stripe mcp to identify any opportunities for how the account is set up and find opportunities to improve
shout out @levelsio for inspo based on his chargeback winback post
🏆 For the first time in a decade on @Stripe I've started winning disputes with my vibe coded dispute responder
I used to ignore disputes so I almost always lost them, now I've started winning, this one is the first big dispute for $1,199 USD!
Whenever a dispute comes in, my site gets a webhook notice from Stripe, it then starts collecting evidence and generates a PDF with entire user's details, when they signed up, and most importantly what they did in the app
In this case the user used the app for months, generated thousands of photos then tried to get the money back from their bank
The evidence has to be REALLY detailed, and REALLY good, which is why it's perfect to vibe code it, you can get quite detailed with different types of users and activity on your app, and put that all in the PDF
I'm shocked because I again I never would win disputes before
People in US especially abuse the [ chargeback ] or [ dispute ] en masse, unlike the rest of the world, it's easily built into their banking app next to every transaction, so it's one tap to get free stuff. And why not? You get free stuff!
It's destructive for business owners like me on many levels, if I get over 1% disputes on my account, I risk getting shutdown permanently by Stripe, Visa and MasterCard, like permanently for life, not just my business but on my personal name too, it's ruthless
Disputes are also super expensive for business owners: you don't just pay back the amount they disputed, for every dispute you pay $30, which you only get back if you win!
But with AI we can now create our own tools to fight back against dispute abuse and finally win! 🎉
Hey Edwin!
I saw someone comment saying they want to use cursor but the discount with cc is so good, and you said “dm me”.
I’m in the same boat! Actually building https://t.co/cbSjmTGCbz which integrates with cloud agents and they are so good! Can you dm me??
I’d love to know what your other chat was about
https://t.co/BrxjoZL4uF
@RampLabs We’ve actually built something quite similar to this that any team can use with https://t.co/lP80yDMKqW
Just connect codebase, set up agents, and auto generate tasks which auto assign to coding agents.
OpenClaw reverse-engineered my local govt’s parking portal to automatically pay parking so I wont get fines any more.
it wrote the script then scheduled it for the time I want to pay for parking.
The script runs automatically, and pings me on telegram when parking starts.
If it fails it loops in the agent to debug and fix, so as long as nothing breaks it wont cost anything to run.
Next step is to connect to Tesla api so it only pays for parking when the car is actually there, then I wont accidentally pay for parking when I'm on holiday 😂
Here’s a first look at X-Plane 12 on Apple Vision Pro!
With visionOS 26.4 and NVIDIA CloudXR 6.0, the simulator streams wirelessly at up to 4K/120fps to your headset.
And if you have a physical yoke or throttle, ARKit uses image detection to recognize them and place them inside your virtual cockpit. 🤯 It’ll be available later this spring.
I ran this prompt over the GroupChat codebase and it claims it built 3 years of code in 8 days, with an estimated $594K worth of value
We spent about 1B tokens and it replaced asana and a bunch of other internal tools we were using, plus allows us to use agents in our workspace.
prompt in comments