Category: Workshop

OK, am definitely feeling the relief this week.  Last Friday I put the final period on the final chapter of ‘The Frequency of Dust‘, my sci-fi novel manuscript.  It’s not a huge one – you can probably kill it in 5-6 hours – but I’m pretty proud of finally wrapping up the first full draft.  Although they may still see some minor tweaks, you can also read the first 4 chapters and sign up to get a ping (no spam- just one email!) when the book is available.

But that’s not all, book #2 I’ve been working on – “Creative AI” for Apress Publishing – also handed off early last week.  Tom Green and I have been hacking at that one for months as well, and it’ll detail how I use AI to augment my creativity… not replace it.  🙂

Check out the books at the links above (or hit the ‘Books‘ link in the site header/footer), sign up to hear more when they drop, and I truly hope you enjoy both of ’em!

Radcap is now at version 1.0.4 – quite a bit has changed since 1.0.2 (public release) so thanks to everyone who submitted their issues and/or feature requests. Specifically, 1.0.4 fixes two annoying bugs with 1.0.3’s new MP4 video format support.

• Recording crash fix — Fixed a crash (SIGABRT) that occurred when recording after switching between video formats (MOV ↔ mp4) while Audio-only Format was set to WAV. Audio settings are now derived from AVFoundation’s recommended defaults for each container, which are always valid and match your microphone’s native sample rate.
• Renamed “Audio Format” → “Audio-only Format” — The label in Settings more clearly reflects that this setting applies to audio-only recordings, not to the audio track in video recordings.

• Video format picker — Choose between MOV and MP4 in Settings → Output. Both use H.264; MP4 improves compatibility with YouTube, Instagram, LinkedIn, and non-Apple players. MOV remains the default.

• Clean recording start – Fixed a ~1.5-second frozen-frame glitch at the beginning of every recording. The teleprompter HUD preview layer now initializes during the countdown, not at the moment recording begins, so the first frame is sharp.
• Menubar recording indicator – The menu bar icon switches to a red recording dot (⏺) while recording is active. Click it once to stop; no menu required.
• Smarter recording start – Duplicate Record clicks during the countdown are now ignored. If the writer can’t create the output file for any reason, an alert explains what went wrong instead of failing silently.
• Capture performance – Cropped recordings reuse the pixel buffer pool instead of allocating a new buffer every frame. Voice-activity detection no longer runs when Radcap is idle.
• Accessibility – Reduce Transparency and Reduce Motion settings are now respected in the teleprompter HUD.

I can’t take credit for cracking the code, but I will take credit for this little app that makes it easy to use.  Apfel is a wonderful OSS code library that unlocks the Neural Engine on Apple’s M-series chips, which normally is locked-down for their own use in Writing Tools, Apple Intelligence, Xcode’s assistant and more.   However, it’s a CLI utility so requires hopping into a terminal to use, so I wrote Grapfel, to wrap around Apfel and give you a friendly way to hit that sweet Neural Engine.

Grapfel clipboard formatting options – Markdown, code or plain text

You can copy output as plain text, Markdown or code blocks (the Neural Engine is surprisingly good at Apple-based technology and coding standards, FYI), and tweak the algorithm’s temperature and system prompt to your liking.  Grapfel also supports attaching small text files to the conversation for limited context window tweaking (but the API doesn’t allow very large files – I’m working on a workaround for this so you can upload bigger papers, research, etc).

No fees, no subscriptions, no security issues (it’s running locally on your own hardware) – Apfel opens up the fun and Grapfel lets you get your hands on it easily.  It only works on Silicon-based Macs (M-series chips) as that’s where the Neural Engine lies, and offered as-is for now.  If you like it and want more features/changes or run into bugs, let me know by logging an issue on GitHub with:

• the specific model of Mac, and version of MacOS you’re running
• what you expected to happen, and what actually happened
• any prompts you were using or working with when the issue occurred
• any crash logs that were generated,or error messages displayed

Have fun playing with Apple’s hardware-based AI model!  I will say that it absolutely sucks at modern/pop culture references, but is great with general advice and AMAZING at Apple technology & development practices.  But you get what you pay for 🙂

I really hate recording social video with crapware social media apps, or even worse – bloated video editing apps that are like bringing a tactical nuclear missile to a fingerpainting session.  So I put together Radcap, a simple Mac OSX menubar utility that gives you a webcam recorder on demand, with a voice-activated teleprompter so you don’t forget that really cool joke you wanted to drop mid-shot.

Given I made Radcap for myself, it only runs on Mac OSX 26 (Tahoe) and greater.  I can’t guarantee a ton of new features or platforms/devices over the next year, but hey- it’s 100% free!
To download or contribute, please visit the Radcap page on my website, or on GitHub if ya nasty.  Log any bugs and/or feature requests that arise on GitHub as well, and by all means – enjoy Radcap!

Claude Design just dropped and oh boy, the rumors of Design’s demise are aflight. To be fair, it’s a great app so far from what I can tell. Should Figma, Canva or Adobe be shaking, tho? Thoughts in the video.

Although it can burn tokens like a campfire, using MCP apps in AI workflows puts the real power of LLMs into the tools you use the most.  In this case, I needed to rebuild a set of CMS views I’d built out in code quickly, and didn’t have reflected in the main Figma file. MCP to the rescue!

Claude + Figma via MCP = complete harmony between my design surface and my codebase… even when I’m working faster than my better judgement.

This is my new band ROADS performing the song Spirits in the Night – a Springsteen tune reimagined in ROADSesque fashion. I didn’t play bass on this recorded track – but will be if you catch us live this year.

I’ve been jamming with the guys for a few months, but it’s now official- I’m joining as their new bass player/background vocalist. I go back thirty-plus years with their drummer, Chris (a virtual human groove box) so the ROADS rhythm section is going to be super tight.  We’ve also been working on filling out 3-4 part harmonies for the background vocals, so that’s been fun.  Enjoy this little slice of Americana and keep posted for more.

So you wanna see (and hear!) what it’s like to fly an FPV drone?   Here’s a lazy saunter through the elementary school next to our condo in Carpinteria California a few years back.  Don’t worry, my son went to the school, we had written permission to fly there on the weekends when no activities were planned, and this was one of those Saturdays.  Dig it, my babies.

So I left my job at the end of September. Without getting into painful details, let’s just say the promises of AI fell far short at my prior employer – both from a technology and expertise perspective – and I found that I prefer to work with AI, not for it.  Or more to the point, I don’t like working for people who have little vision or experience beyond what AI can provide them.

OK, ouch.

But to be honest, that’s being awful kind given the reality of my former manager’s incompetence.
So sorry, not sorry too.

I really needed to revisit what innovation and hard work meant in a world where AI is nipping at the heels of every job, every role, every industry.  And what experience and seasoning means in a world that now generally looks for their advice from a synthetic LLM.

I had to look back at a career spanning some of the most influential companies in the tech and design space. Having held key roles at LinkedIn, Lynda.com, Adobe, Macromedia, Metacreations, and HSC Software, I’d originally built a reputation for combining technical expertise with innovative design thinking. My work across these differing companies and markets has given me a unique perspective on digital experiences, content strategies, and creative tools, and ABSOLUTELY shaped my approach to solving complex challenges.

And that’s what I do best.  I break through limits and constraints, and find creative solutions to tricky problems with a creative, empathic eye.  AI can do the first part very well.  But it will never do the second part in a truly human way.  And when I used AI, I didn’t look at it as a force eliminator, but a force multiplier.

So today, I’m the principle at Brilliant Mindworks, a dynamic and scrappy creative consultancy I founded up here in the Sierra Foothills of California.  Now to be fair, it’s really just myself and a few teams of AI agents to let me focus on the important and creative parts of my work, and most all of my projects are remote and international.  But I have friends, partners and colleagues in about every corner of the world, and of the creative and tech industries – ready to help when a problem exceeds my capacity.

Brilliant Mindworks specializes in strategic projects of all kinds, from digital product experiences to brand-focused initiatives.  I like to focus on high-impact solutions for clients who value creativity and precision in their products, campaigns or projects. If you’re looking for innovative and thoughtful design strategies that truly elevate a plan, I’m your guy and we’re your agency.  Click around and learn more.