Daily log

unlearning 20 years of building product to become a “native ai builder”

things i’m trying and really enjoying

• building a cloud based agentic workforce.. goal is to reach or exhaust 1b high quality tokens per day
• design.. entering peak play era 😍
• building context graphs
• wide deep dive research
• talking to new non-technical builders who are making magic for the first time
• revisiting old projects, products, and problem statements

things i’m failing at

• context debt (project-based)
• unblocking runners and keeping agents in flight (peak 😤 when I see runners in idle for 12 hours)
• saying no to new features or products (esp during design explorations)
• keeping token spend balanced throughout the day and ramping output during off-peak hours

rejected pill labels from figma, and claude trying to attempt handwriting a new logo.. wip:

ai kills the idea maze and the stress tests that happen in the 0 to 1 process

the idea maze gets built and refined subconsciously when you write the prd, the deck, the 1 page north star doc, the wireframe, the db schema etc

when you skip those, you sometimes you get lucky.. a lot of the time ai takes you down the wrong maze

It feels a bit like we’re entering an era of the “0 → 0 paradox”

Jumping straight into code can feel like you’re pushing from 0 → 1 faster, but the lack of design thinking can set you further back than where you started

Modals everywhere, complex flows, overly/under branded…

— luis. (@disco_lu) October 14, 2026

it’s Thanksgiving in Canada and the sun is shining bright.

PSA: the sprint is on, just 70ish days left in the year.

shifting this log towards ships and lessons learned. blah blah.. that’s fewer posts, more building..

it’s Friday, 8:30pm. just closed 50 tabs, and it feels wonderful!

shipped four tiny websites and launched fifty plus deep dive research projects. these LLMs paired with agents are truly an analyst’s, researcher’s, PM’s, or curiosity cat’s Game Genie. these are remarkable and extremely exciting times.

three things top of mind:

1. my little builder thesis continues to take shape. it’s no longer little and ballooned to 200 slides yesterday. managed to get that down to a hundred. i have 20 new slides to add, and from there i’m hoping to prune the entire thing down to maybe 60. but let’s be real, who has time to read 60 slides? guessing a lot of people will just pump this through a summary agent. hopefully, they’ll take the time, and mega thanks to peers and founder friends that are giving feedback, poking holes, and casting a little bit of shade. hahaha
2. how much to share on this daily log
3. how to showcase previous and ongoing work. the barrier to replicate, especially for an early stage project, is near null. all of the interface work, the builds, anything before 2023, is it even worth showcasing? even the products that we’re developing today aren’t AI or super intelligence native. we haven’t even scratched the surface of what products will look like, feel like, and the infrastructure is laughable.. or we’ll look back and laugh at how we were trying to build AI products on a web and mobile world. there’s new gravity duh

how many deep dive research projects are you kicking off daily? am i the only one running more than 10 per day?

How many brands and startups that you previously sunset have you restarted? The temptation is real, and i’m bringing back at least one (for now).

Late last night and early this morning I created hundreds of logo variations for a restart (that I’m keeping under wraps).

wip

strat is starting with a monthly recurring service that turns into robust agentic systems from there. let’s see

VIFF is almost wrapped and I’ve thoroughly enjoyed all three films.

Tonight i’m deep within a thesis project that’s taken me way too long. fingers crossed I wrap it tonight

Google Alerts but for your own ideas being market ready

need some sort of deep dive research vault. easily spent more than 50% of my tokens this week on research and PRDs.. that ended up going abso nowhere.. killing them at PRD and DD stage.. which is all good, but the shelf life on industry and new idea SWOTs and competitive DD is shrinking from a year thing, to months, to weeks..

so this vault needs to schedule re-checks and trigger a ‘market is ready’ alert

new possibilities and universes to explore.. let’s goooo

ahhh or argggg, back inside Figma. After giving keys to Fable I’ve reclaimed those back, working through what looks like the third rewrite of my thesis deck.

so far my favourite aspects of working through decks with Fable and Figma slides are the ability to produce certain designs or specific layouts and have those recreated across huge decks =)

Since this is a v1 deck without a brand it’s taking a bit longer.. however, I plan on loading this into claude design system for the next one

My least favourite part is Claude producing designs on its own and any sort of storytelling or flow is just incoherent…

The thing about a 20 slide deck previously taking hours to construct vs having AI now do it in minutes is that the hours it took to create it in the old world meant that everything was thoughtful, crafted, and scrutinized. In the hands of someone good, it made for breakthrough…

— Michael J. Miraflor (@michaelmiraflor) September 26, 2026

and back into figma slides with ANOTR =)