#bitcoin / cofounder+cto @getjustai (yc w24) / ex @automattic @twitter @mercari_inc / @stanford cs + phil ('18)
San Francisco, CA
Joined January 2019
- Tweets827
- Following2K
- Followers627
- Likes1.1K
Pinned Tweet
got interviewed at @TechCrunch about why a bunch of ex-tweeps and i got together and started @getjustai to disrupt marketing!
can’t believe it’s been almost exactly three years since @elonmusk took over @X, but i probably never would have had the guts to try or the endurance to keep going if it wasn’t for that experience.
some random lore, geohot sat across from me during his “internship,” and he mostly just talked shit all day haha.
other random lore, i was in a conference call with elon, and when we went around the room to introduce ourselves, i flubbed my own intro because i got nervous and elon ended the call right after. i thought i was gonna get canned over that 🤣
but anyways, the first year of doing a startup was the worst, and it’s extremely humbling to grind with no funding, no salary, and no revenue for months. incredibly grateful to be working with amazing companies like @NotionHQ @coursera @figma and to partner with marketing leaders like @Iterable @Braze and @CustomerIO.
shout-out to the #TwitterMafia, or the all ex-tweeps who started building startups too @neha01mittal @ajeet @paraga @iuliabaltoi @yolandaaacao
reintroducing JustAI… it’s cleaner.
it’s been almost two years since we started our journey as Just Words AI, and it feels like as good a time as any to announce our rebrand to @getjustai
it’s an exciting time for us — just this month, we just hit a Major revenue goal, closed some deals with amazing companies, hired our third engineer, and next week, we will be pitching at @TechCrunch disrupt.
there’s a lot to a name, and what i always liked about “just words” is that it’s a bit tongue-in-cheek. in every sales pitch, we’d promise huge gains from copy changes, and i liked the double entendre that we can do it even if they’re just words… or maybe we’re just words. and to be honest, we did have to eat our words sometimes, but every time we failed, we had to adapt. we’ve had to evolve so much just to survive, and i’m proud to say that we are a different company today.
late to the game, but liking formal verification as a way to file bugs / define specs for how services should interface with each other
the meeting note taking apps should really just be replaced with a meeting agent that actually participates in the conversation
went to a cool conference in Austin for AI experimentation and the one picture they got of me was being a fuckin nerd
this was a long journey of our own where we had to shift from a product that was working and making money into a platform that stands on its own and supports enterprise-grade scale.
Neha and I had to have extremely tough conversations about where we saw ourselves against where the industry was heading, and we decided to take the hardest option - to start over from first principles and build everything from the ground up.
We took in all of the things we’d seen and experienced first-hand inside enterprise marketing teams over 3 years, grew a cracked engineering team and a mean-lean GTM team, and blitzed building an end-to-end marketing platform to compete head-on with 10+ year old incumbents.
easily the hardest engineering feat I’ve ever done, and it started at one of our lowest points, where we just decided to stop coding for one week, to whiteboard and sketch out a new architecture, and to fold paper cranes while we bounced ideas off each other.
I should really write an engineering blog about how we’re using durable objects and the mistakes we made along the way.
📣 The wait is over. We've been teasing October 19 for weeks, and today we're finally showing you what we've been building.
Meet the new JustAI: your entire marketing workflow, powered by one Marketing Harness. 🚀
getjust.ai/launch?utm_source…
“you can see everything” was like a real-life vertigo.
Nathan fielder was like an autistic Jimmy Stewart to Elizabeth Holmes as Kim Novak. The vintage feel, the violin scores, the general creepiness of being overly intimate in following them around, complete with the Judy character who seemingly gets attracted to Nathan who is obsessed with Madeleine.
at Twitter, I guess we could have stopped people from tweeting at New Year’s Eve and the superbowl at peak hours to smooth the demand curve
hey @thsottiaux , this morning half of my chats are grayed out and inaccessible.
if i hover over them, i see a message "Unexpected server response" and clicking "Reconnect" doesn't do anything.
kchro retweeted
Introducing Replicas V3: the most flexible and cost-efficient cloud agent on the market.
- Run Claude Code, Codex inside VMs
- Bring any subscription or API keys
- Delegate from Slack, Linear, desktop app, or your phone
Now proactive, and more powerful than ever.
replicas.dev
octopus 5.5
For years, lifecycle marketers have described their journeys the same way: an octopus.
So we headed into the tall grass and caught one. 🐙
Stay tuned. Your journeys are about to get a whole lot easier.
#JustAI #LifecycleMarketing #Opus
the optimal strategy at this point is to buy all the subscriptions, blast through weekly usages as fast as possible, and rotate through the subscriptions
kind of like churning credit cards
Stop stealing our Codex quota with these surprise hard resets!!!
Not everyone burns 100% of their usage immediately and waits for a reset. Some of us actually plan our quota around real work. (& have multiple AI providers or accs)
I had ~60% usage left and A surprise reset means losing that saved usage nd pushing the next reset another 7 days away.
That’s not a free reset for everyone right?? for some it’s literally less usage.
At least give us an opt outor banked reset. @thsottiaux
It's pencils down, people. Writing code by hand is no longer an economically viable skill for most programmers at most companies. But the future of making software has never been brighter. Don't you dare black pill this beautiful moment! youtu.be/vDjW_dRyKXY?si=6Fsf…
interesting that this post has inspired a lot of (big) companies to take the leap... the "low-risk change?" decision block is the biggest AI pill to swallow
Here's what OpenAI's agentic software factory looks like, today.
Details: newsletter.pragmaticengineer…
(thanks to all the OpenAI folks who explained how it works! And Perf Factory looks especially interesting to me)
