Special thanks to Arthur Sabintsev for the feedback and review of this post.
Most people overestimate what they can do in one year and underestimate what they can do in ten years. - Bill Gates
You have to live it to understand it, and this statement can only come from experience. The one thing Bill forgot to mention is how much will change in the span of 12 months within each of those 10 years.
Turning 30
Having turned 30 this past year, it was a cause for self-reflection, as I’m sure it is for many others.
10 years ago, I was finishing up my first semester specializing in Computer Engineering at UofT after an iOS summer internship at a 500-person Twitter company. Since then, I’ve worked across more disciplines than I ever expected to touch in my life, gone from knowing nothing to something in topics that still blow my mind, have spoken with people I never thought I’d meet, and revamped the timelines of my financial aspirations. I recovered from more physical injuries anyone should ever encounter, while falling behind in my personal life relative to where I thought I’d be. I have grown as a person alongside some of my best friends, slowly grown apart from others, and met new ones through unexpected means.
One of my favourite reflections is being proven wrong by my sister in this letter, as she is about to finish her first semester as a Speech Therapist at Manchester University.
A Year at Pocket
Just about a year ago, I wrote about my life-changing trip to Mexico, which is how I ended up officially joining the Protocol team at Pocket on December 1st, 2021.
I envisioned that by the time I hit my cliff vest date, I’d be working closely with Luis (CTO), Andrew (Founding Engineer) and the rest of the protocol team in improving and maintaining a live & functional Pocket V1 protocol. I expected that a few people would join our small and scrappy team to deliver on the huge vision we all set out to build. I expected that we’d be looking back on a year of successful R&D cycles while collaborating over a digital whiteboard to figure out what’s next. I expected a quaint, fun and HR-compliant celebration.
I did not expect that I would have just returned from Pocket’s SOTU (State Of The Union) in Tampa, where core team members, external contributors, DAO members and investors gathered to present, listen and discuss the past, present and future of a 75+ core Pocket team, with orders more of external contributors. I did not expect that chaotic drive would turn into a structured passion in an org led by Arthur (COO) and Laxman (CFO). I did not expect Luis would have transitioned out of his position as CTO, and I did not expect that Andrew would be handing things off just a few months later. I did not expect to be managing the protocol team alongside Jess within an org that is just as much product-led as it is eng-driven. I did expect a quaint, fun and HR-compliant celebration.
Moving, Fast & Slow
I can confidently speak for every engineer out there when I say that we’ve all had a sprint, be it a day, a week or a month, where we got so much done it becomes a prominent lifelong memory. I’ll never forget that time @amandeep, @davyli and I were literally lying on the floor of the Toronto subway, debugging our robot, as we were heading downtown to demo the final result of the Atomic Drum Detector for our AER201 course. As invigorating as these moments are, they’re very bad at re-calibrating our personal expectations of what we can achieve within a certain time period on a recurring basis.
Daniel Kahneman’s Thinking, Fast and Slow is a treasure trove of knowledge, but it’s one of the few books I believe is done justice by its Blinkist summary. The tl;dr is that individuals have two modes of thinking: instinctive (thinking fast) and logical (thinking slow). Similarly, working at a startup comes in two modes of operation: sprinting (moving fast) and jogging (moving slow).
I spent the first 6 months at Pocket sprinting, and these were some of the most productive months I’ve had in my career. The last 6 months were spent jogging, with outputs that are somewhat less tangible. Part of it was research, part of it was work, part of it was planning & preparation, and though it was infiltrated with “distractions”, part of it was just being a human being.
In the past 6 months, I hosted a panel of crypto OGs at Infracon on “The Demand Side of Protocols,” gave a lightning talk at The Science of Blockchain conference introducing The RPC Trilemma to the world and planted a bunch of seeds in my research on Merkle Trees that are still germinating. I also wracked my brain for a few weeks as to whether a custom in-house built blockchain is really the right way to go after absorbing more information than I could handle attending Devcon; spoiler for a future talk: it is.
Behind the scenes, the entire organization was restructured, the team took a few different shapes, and we started to really define how to make our processes & operations both constructive & effective.
Over the years, I’ve been at slow research-driven organizations, fast mobile-first companies, and top-down management structures that aren’t quite sure what velocity they’re trying to move at. Pocket is a tech-driven movement where both the product and the engineering need to balance each other out so we can move, both fast & slow. Pocket is still small enough to adapt quickly but big enough that there’s no single voice (internal or external) paving the entire path of the project. Within the core team, Arthur structured the three technical pillars (supply, demand, protocol) via a buddy system that is co-led by both product and eng. Though we’re still figuring out some of the kinks, I’m constantly exhilarated during moments of cross-team alignment and euphoric moments of ideation where product meets eng and eng meets product. As this continues to improve, I genuinely believe it’s the most optimal design at this stage of the company, enabling us to move both fast & slow.
One of the best things that happened to the Protocol team in the past 6 months is having @jalde__ join as our PM. I’m not sure if the appropriate title is Product, Project or Program Manager, so it’s fortunate that Protocol Manager fits the acronym of PM just as well. It’s only been a few months, but I’ve heard people say that crypto operates at 168 hours a week (24/7), so it already feels like ages of having a partner in crime. Roadmaps are aligning, structures are being put in place, and velocity is slowly increasing, allowing us to move both fast & slow.
On the technical side, one of my personal favourite changes on the Protocol team is starting to leverage Github discussions for longer-term ongoing research-driven conversations while keeping most of the day-to-day technical back & forths almost entirely in public via GitHub PRs and Issues. I’m also excited to start leveraging Dework more on the protocol side to get the community involved in tasks I’d classify as “Important But Not Urgent” per the Eisenhower Matrix. I’m envisioning interoperability, zk-SNARK approaches to volume validation, applying BFT forensics research, and large-scale DKG, but I’ll stop here before I get too off-track.
Oh yea, it’s worth mentioning that the whole company and community came together to solve a Chain Halt. The root cause was found relatively fast, but applying it to a distributed network was slow. Though it’s no badge of honour, I think it speaks to the fact even though finding the root cause was on par with any bug in a traditional company, our validator set is decentralized enough that we can’t just restart all the nodes ourselves.
The Crypto industry moves fast, but protocol development is slow. Developing a prototype is quick, but its validation is slow. Ideas can come fast or slow, but the speed of decision-making depends on the context. As the saying goes, “slow is smooth, smooth is fast,” and as we continue smoothing out the edges, everyone part of our process, from research to ideation, then decision-making, validation and verification, will be both smooth and fast.
The development of V1 has undergone a lot of changes in the past year, causing some iterations to move fast while others moved slowly. However, the team is quickly taking shape, and we have several new people who will be joining in the near future. I’m extremely excited to see what Pocket’s Pulse will look like around the time of my next post.
Self Reflection
Growing up, I competed in over half a dozen sports. Though I was usually half decent, I was never a pro. Similarly, I’m learning to accept the same of my technical abilities. Unless I fully devote myself to one thing only and nothing else, winning The Levchin Prize is probably off the table, but being well-rounded enough to help bring the Protocol team to where it needs to go is within reach.
Whenever I’m tackling a new problem, technical or not, I almost never have an immediate answer. I often need to sit and meditate on it for a little while, letting it oscillate from my conscious to my subconscious as I go about doing other tasks. I need to fully immerse myself in the problem. I read and listen to relevant materials, discuss it with those around me, and then take a step back to let it simmer for a bit, repeating the process until a solution pops up. Unfortunately, I don’t have a Hyperbolic Time Chamber so the rest of the world can wait. This is why a team needs to be able to simultaneously move both fast & slow. The balance of measuring twice and cutting once, while not taking too long to measure is a fine one, and we’re still working on optimizing it.
In a conversation I had with my sister on a related topic, I got this response:
Though I still struggle to understand why Gen Zs can’t spell out “You aren’t,” I believe the statement holds true. I’m either all in or I’m out. Right now, I’m all in.
No FOMO
In a series of spontaneous Twitter threads, some Pocket community and team members all participated in taking the Myers Briggs Test together.
I compared my results from 2017 (INTJ-T) to 2022 (INTJ-A).
As an Introvert, I don’t often experience FOMO at social events, but do have a Fear Of Missing Out on opportunities and challenges in corners where others aren’t looking. When everyone is focused on Data Availability, I think about how we can improve Data Redundancy. When everyone is focused on high TPS and short block times, I think about large block times and increasing the validator sets. When everyone is focused on modular blockchains, I think about optimizing for a specific use case. When everyone is discussing node running, I can’t shut up about light clients.
I’d call it a form of reverse FOMO where I’m afraid I’ll be pulled in with the crowd and miss gems of hidden opportunities. This could be a recipe for innovation but also one for getting lost in the woods. This is where having a strong team, a vocal community, balancing product & eng, while being able to move both fast & slow is important.
One other avenue where I succumb to FOMO is getting into challenging and meaty conversations with people much smarter than myself. Luckily, working on an open source project and having the privilege of travelling to attend various events, I’ve had challenging conversations with people of all ages, from middle school to retirement, and literally all around the world. In addition, long-form podcasts like The Lex Fridman Podcast, The ZKPodcast and The Tim Ferris Show are opportunities for me to be a fly-on-the-wall and feel like I’m sitting at the table with the most brilliant minds of our day, without needing to leave the comfort of my own gym.
The Last Year
The last year has been all about figuring out what we’re building, who’ll be building it, how we’re building it, why we’re building it and setting the appropriate expectations. We’re scoping down our features but not our ambitions. The next year is all about staying focused, executing and setting the foundation for the next 10.
This past year was also the first that Pocket saw competitors enter the market as the letters R-P-C started appearing everywhere in the Web3 infrastructure space. People say that imitation is the highest form of flattery, but I also see it as an opportunity to double down on one of crypto’s core uncovered values: coopetition.
In the recently published book, The Power Law by Sebastian Mallaby, he speaks of how coopetition was one of the key drivers to Silicon Valley’s growth & success in an era primarily driven by Type 2 Volkswagen vans. With Silicon Valley losing its luster, the (real) Crypto industry has an opportunity far to exceed the coopetition of Silicon Valley circa 1980. When else has there ever been this scale of both collaboration and competition in a transparent, digital and global world of builders?
I may sound like a broken record for those who work with me on a day-to-day basis, but I still struggle to fully wrap my head around Pocket’s opportunity ahead. Building a strong community and ecosystem around an application-specific blockchain is far more powerful than a general-purpose blockchain without the community’s building blocks. Chaining those together enables us to avoid focusing on a general-purpose storage layer or execution environment (i.e. VMs or smart contractors), but rather hone in and optimize on a solution to the RPC Trilemma. We don’t have to stop at just building a blockchain that works, but we can prove and explain to anyone how and why it works. We don’t have to build new features ourselves but rather provide the tooling to empower others. Most importantly, we're already doing it:
I’ve personally started exploring Celestia’s Sparse Merkle Tree and planning on adding a visualization library so we can see what’s actually in the tree; celestiaorg/smt/pull/74.
@germany_beal has taken on one of our bounty tasks to build a visualization of our RainTree algorithm for structured gossip. You can see an early prototype here, and we’ll release more details soon!
With so much going on in the last year, the entire leadership team under Mike has changed. Though I knew the departure of both Luis and Andrew was inevitable, it did catch me by surprise. I thought we’d have more time to work together and learn from each, but I’ll cherish what I got. You’re never ready to be thrown off a cliff, but when you are, you have the option to either fall or soar. The latter is much easier when you’re not doing it alone.
It’s a whole new team, a whole new plan, a whole new structure, moving forward with the same ol’ vision, simultaneously both fast & slow. The community continues to grow, and core team members hand off the torch with pride so the flame doesn’t go out. V1 is Pocket’s 2nd mover advantage, where Pocket itself is also the 1st mover.
The Next 10 Years?
If the last year is any sign, a lot will probably change again over the next one. The next 10? Only time will tell. I’ve got a pretty long list of personal goals I want Pocket to hit, but also plenty of other ambitions that may take multiple decades to play out.