My fellow Mavys, I have loved working with you these past 5 years. You’re thoughtful and caring and silly and hard working and brilliant. You’ve put up with my 11/10 energy, and you’ve made many 10/10 moments happen. Thank you.
Here are a few parting thoughts, things I’ve tried to do, things I’ve learned along the way, and things on my mind as I start Book 3 of my life.
Make breakthroughs1
My favorite moments of building were when I’d get excited about an idea and pull a demo together in a few days. It would almost always be outside my core work, I’d work nights and focus intensely on it, and sometimes I’d prove something that made a difference to the business. It made for some fun demos at least.
This, to me, is the great gift of being a software engineer. To know the problems and the tools and to see a path and to have the power to make it real. It’s almost necessary for this to be permissionless at first, before theoretical reasons get in the way of practical possibilities. More than anything below, this is my wish for myself, and my wish for you.
Take time
Gagan reassured me early in my time here that it wasn’t just ok, but actually good to have big blocks of time that were not spoken for, and that creative work required space for thinking, writing, and generating ideas.
I hope in this era of background agents we can take the time that we save and use it to think more deeply. To move on from things a little slower. To pair shorter build times with longer discussions, sharper ideas, bigger plans.
Talk more
Most of our work revolves around decision making, and I’ve found that decisions are much easier to make in conversation with someone, articulating hazy thoughts, asking and answering questions.
I have rarely regretted (and have often regretted not) talking more with collaborators, talking for longer than the scheduled time, talking sooner than a calendar would suggest2, and even talking out loud to myself. The sooner the next conversation happens, the sooner the next decision gets made.
Be useful, not clever3
I worked with a lot of smart people at Google, but there was a big difference between the ones aiming to be clever, and the ones aiming to be useful. The clever ones had quick comebacks and ready reasons against new ideas. The useful ones were slower but more optimistic and found creative ways around the roadblocks. The clever ones often held us back. The useful ones usually pushed us forward.
The second best time is now
It’s been noted that I’m willing to ask tough questions even at very inconvenient times. I’ve worked hard at this. To appear ignorant, to be wrong, to be difficult, in service of the idea, in search of some truth.
Acting with courage is my most cherished leadership (and life) principle. And I remind myself all the time that if the best time to ask the important question was earlier in the project, before work started, the second best time is probably now.
AI changes more than we think4
This may be obvious but it’s easy for a change this monumental to be under-appreciated. AI has changed so much so fast, and will change a lot more very soon.
The way we work can and should look radically different in a few months. Pods can shrink, pairs can do more, loops can be automated, ideas can take shape sooner, dreams can get bigger. All this and more, if we break our patterns and reach for it.
Stay silly, stay gold
The thing I love the most about you lot is your ability to mix the serious and the light, the sacred and the profane, the high and the low. I hope you’ll keep this spirit alive. I believe you will.
Love,
Shreyans
January 30, 2026
It’s a time of change, for the world and for me, and it feels like a great time to document the things I’m making and learning. Startups, art, technology, fatherhood, and more. Always hand written, always open minded, usually optimistic. Thanks for reading.