busymichael
Joined 9 years ago
Karma
87
Submissions
99
Member since
Jun 1
https://basilai.app
View on Hacker NewsRecent submissions
Comment · 3 months ago
I think you're underweighting the Pelican test.
Not only does it give you a super easy-to-grok understanding of the model quality just by looking at the image, but when you compare tokens and costs (both input and output), you really get a good, simple COST x QUALITY evaluation across models.
Simon explains it well: https://simonwillison.net/2026/Jul/16/kimi-k3/#what-can-we-l...
Simon, you should put up a summary table page that you update after every release.
Comment · 4 months ago
Still no 256GB or 512GB Studio models at any price. 96 is the max for any Studio configuration on apple.com right now.
Comment · 4 months ago
I built an iOS app for my sons who play club basketball. It’s designed to bridge the gap between practicing solo on the court and the pressure of playing in a game.
It uses mechanics and body movement, via visual detection through the cameras, to identify when you’re about to take a shot or make a pass.
It then pumps distracting noises (honks, cheers, jeers, etc.) into the headphones to try to distract you. It also runs a continuous 24-second clock so you build a natural sense of how much time is left before you need to get a shot off.
I'm looking for more people to join the test flight: https://clutchshot.app/
Comment · 5 months ago
Building Basil (https://basilai.app), a privacy-first AI executive assistant / note taker. Everything is on-device: transcription, summarization, etc. So nothing is shared with any model provider.
Lately the interesting work has been less about raw transcription and more about making the output actually useful: recurring workflows, follow-ups, and personalized summaries that fit how someone runs their day. We are also pushing more toward reliable agent automation instead of one-off chat. And dealing with all the iOS limitations on background work.
Still early.
Comment · 7 months ago
Building a voice recorder/transcription/summarization app where everything runs on-device. No cloud, no uploads, no servers processing your audio.
Started with a web version using WebAssembly Whisper to validate the idea. Worked well enough to prove demand, but browser sandbox limits (no background recording, inconsistent model performance) pushed me to go native.
Rebuilt as an iOS/macOS app with CoreML. Apple Silicon handles Whisper-class models locally without issues, but its confined to recent devices.
I'm working on e2ee sync between web and mobile. I want to build a simplenotes but for audio transcription.
Try the web version at: https://basilai.app/app
Comment · 1 year ago
My experience share only supporting the latest OS:
I have launched apps focused on a new feature in the latest OS and regretted it. The # of people who have the latest OS is much smaller than the full install base for much longer than I thought. As a result, my marketing conversion was unnaturally low - people who liked the app idea but couldn't install because they had the wrong OS. This causes two problems: potential users I activated but couldn't convert and this signal gets internalized by the App Store, pushing down future impressions.
Now I always have a fallback implementation of the feature so I can target the prior OS. Both Mac and iOS.
Comment · 1 year ago
As a side project, I just launched a privacy-first web-based meeting transcriber (https://basilai.app/app). Everything runs entirely in your browser — both the transcription and AI summarization — so no audio or text ever leaves your device.
I'm using the browser built in transcription service plus downloading a model and running it via webgpu. No login. At the end of your meeting, you get a zip file with the audio, transcript and summary.
Comment · 2 years ago
I'm a big fan of digitpain (Jeffrey Scudder) and it is great to see him get airtime here. The tl/dr for Jeffrey is he make digital art, but first he makes his own digital tools to make that artwork.
For those that are new to him, be sure to browse to the top level (https://aesthetic.computer) and play around in the 'terminal'.
If you have a VR device, view Freaky Flowers in it.
Comment · 2 years ago
The feature I like the most in your demo: the one page checkout. It comes very close to shopify's excellent check out. Nicely done!
WooCommerce should be filling the role of viable self host alternative to Shopify, but it falls short. Even with the plugin ecosystem, it is much harder to run an quality cart and checkout.
Congrats on your launch!
Comment · 2 years ago
It looks like you still have rrweb running. Have you tried removing that?
Comment · 3 years ago
I am on a similar path with https://intro.wtf (AI product merchandising for online retailers). I am solo but am trying to make being a solo saas founder the 'water I swim in.' I am connecting with others in my network who are solo founders, joined microconf, listening to podcasts, etc. I am trying to connect with other people who are all focused on similar goals, combine that with my focus and work to get better over time.
Comment · 3 years ago
I'm working on building better product merchandising with AI/ML for online retailers.
The problem: creating unique, engaging product content is expensive and time consuming for retailers. Hiring content writers can cost $50-100 per product. Multiply that by 1000s of products and then add in constant product churn.
With uniquely tuned AI models, combined with a retailers raw product data, I can generate content that improves the buyer's experience and leads to:
- higher conversions
- improved SEO
- standing out from competitors
The app includes:
- the unique LL models
- pipelines to connect to the retailer's catalog
- a feedback mechanism to monitor the products (add to cart, conversions, on page time, etc) and test new content
I did pilot with a large wine retailer. We generated nearly 4,000 product descriptions and saw a near immediate increase in on page time and conversions. It is 90 days out we are now seeing an SEO gain too.