Chocolate, typescript, art, book worm 🏳️🌈
San Diego, CA
Joined February 2009
- Tweets8.1K
- Following265
- Followers814
- Likes13.4K
case retweeted
If you want to know how @linear’s sync engine works, Evan Hu reverse engineered it and did a write up that is probably the best documentation that exists - internally or externally: github.com/wzhudev/reverse-l…
case retweeted
“Build something people want” is such a great mantra for @ycombinator.
Instead of only focusing on “build” and “something”, I wish more builders would also think about “people.”
Human nature is the invariable anchoring all businesses, and it hasn’t changed since we left the plains of Africa.
case retweeted
It's happening! I just used my credit card to pay for the bus!
MTS has entered the final testing phases of contactless payments ahead of the launch in the next week or so.
case retweeted
The default is no one is willing to tell you that your work sucks. And you go on sucking, forever, feebly attempting to reach your goals but with insane amounts of slack in the system, but thinking you’re doing great.
To me, this is sad. This is delusional. This is the thing we should all be afraid of.
I love when people rip my work apart with useful feedback.
As you get more senior and more skilled at your craft, it becomes harder to find people who are better than you at your craft. And the people who are better than you are often quite busy or unavailable, so it's hard to get them to rip your work apart even if you wanted them to.
If someone is giving me detailed edits and they have the skill to tear my work apart, it means they’re better than me in some way and I am lucky they care enough to help me.
If you have someone like this in your work--a manager, coworker, friend--take the time to thank them and consider yourself fortunate.
case retweeted
Early-stage product development is no longer about validating novel ideas for new markets. It's about reaching an ever-higher quality bar to compete in crowded categories.
We need to rethink the concept of the MVP.
→ linear.app/blog/rethinking-t…
case retweeted
I made this guidebook for my 1-1 manager meetings with reports.
What would you change or improve?
case retweeted
Ok guys hear me out, what if we move Silicon Valley to San Diego.
Best quality of life in US, already pulls 3x capital than Miami. 1.5h flight to SF, 1.5h drive to LA
case retweeted
Hosting an ice cream social in San Diego at Yogurt World UCSD (9500 Gilman Dr, La Jolla) on Thursday (2/8) from 2:30 - 4:30! See you there!
case retweeted
We are opening a software engineering position at @reflectnotes. If you want to:
1. work with the latest web tech
2. on a product loved by thousands
3. with complete autonomy and only one meeting a week
Then this is your chance. We typically only hire about one person a year.
case retweeted
People really over complicate companies with OKRs, 1:1s, status updates, roadmaps, blah blah blah.
We don’t do any of that at Reflect. We have one all hands a week and a WhatsApp channel. That’s it.
And yet we ship lots, have >4k customers, and growing faster than ever.
case retweeted
I want to start a factory nerd tour group:
- Regional chapters
- Arrange monthly insider tours of cool factories via the connections of the folks in the group
- Get food/beers after, since manufacturing nerds are the most fun to hang out with
- Annual overseas trip to tour an itinerary of factories, think Japan, Germany, etc
Not quite sure how to get this started, but it would be very fun.
case retweeted
Many told me that work trials won't work or won't scale.
Over the last 4 years, we hired 50+ people with work trials and with a 96% retention rate. Everyone from engineers to C-level candidates goes through one.
The best way for both parties to test if there is a fit.
How do you identify the right people to hire?
At Linear, every hire goes through a work trial. In this new post, we share the thinking and the mechanics behind them.
→ linear.app/blog/why-and-how-…
case retweeted
Founders, one high-impact thing you can implement now: ask your users where they heard about you in your signup flow.
This data has been pivotal in our growth. It tells me where our growth is coming from so we can double down. It surfaces partners we should work with.
case retweeted
We're doing something different with support, and how we work, at OpenAI. Here's an excerpt from a note I sent @bradlightcap before joining:
If this resonates with you, I'd like to meet:
* User Ops: boards.greenhouse.io/openai/…
* Engineering: openai.com/careers/software-…
case retweeted
Hearing @linear 's designers take screenshots and designing on top of that, disregard design systems and fancy figma files is a refreshing revelation.
"Because in the end, what is the user looking at?"
He's so right. We should focus on work that improves the product.
Obviously this type of productivity is best suited for small teams or solo designers. As you scale, it's a good idea to get organized around a design system.
case retweeted
✨ Announcing Reflect ✨ – A new way to build multiplayer apps like Figma or Notion.
Rather than CRDTs, Reflect syncs the way video games do. Today, Reflect is available to everyone.
Learn more: What the Multiplayer Web can Learn from Video Games ⏩
rocicorp.dev/blog/ready-play…
case retweeted
To me the fascinating thing is that a lot things work or doesn't work because of the setup or history of the company.
It's bit like people would say that jumping out of airplanes doesn’t work but then didn't pack a parachute when going up.
Couple of things I think worked well for @linear:
- Very clear focus on the top customer from day one. Before starting, we probably spent a year researching the space, problems and refining the ideas.
- We ran a 1-year invite private beta to focus getting feedback from the defined customer profile & the companies that could use it as we were building the product
- Thanks to @artman we had very well thought out technical foundation and engineering fundamentals. We haven't needed to time untangling a mess or rewriting a lot of things
- The sync engine pushes the technical foundation further because product engineers don’t have to think about managing data. Features are faster to build. Last 4 years have shipped close to 170 changelogs (usually including one larger update + 10-20 small fixes)
- Have a community and shared Slack channels with customers. Have everyone in the team join and answer questions which helps them build understanding of customer problems
- Have product teams to have space and time to explore options and polish the experience. Push deadlines back if needed
- Being remote avoided most of the chaos of pandemic, and now the tug of war of in-office-hybrid-remote. Also as founder I don't have to spend time thinking about office furniture
- Hiring slowly, and senior people in all roles, including support. Only hire if we are “strong yes” and not compromising our values. Our rule has been not more than double the team at any given year. Any new function or role, always hire one or two people first. Get the learnings and scale only after something works.
- Using worktrials to validate people's skills and how they approach problems, avoiding wrong fits.
- Hire leadership who lead from the front and are willing to roll up their sleeves. Every person in Linear has to be an individual contributor as well. Get output from the day one
- Not a lot of meetings. Try to do async, video recordings, shared meeting records.
- Focus on the brand from day one. Brand is about communicating your values by the actions, words and visuals
