What is the ride?
For me the ride has always been either:
- Staying in the US (getting good at 1v1 dueling USCIS examiners in Nebraska like it’s a JPRG), and building a life here!
- Getting to the day where I can have another attempt to build a startup after the failure of the last.
For me the ride used to be about graduating law school[1], and somehow erasing my shyness so I wouldn’t get eaten up by big law. But that was probably insanely doomed. And besides, while I would crush it now after going through the demands from my last startup - it seems really boring in retrospect that, that was going to be the ride I chose to give my time to.
Getting off that doomed ride, and getting onto this one is exactly the reason this post exists - so let’s get a-going.
This post is long, but it’s fun. It documents in short, most, if not all the decisions and experiences that lead me to where I am today. I share this mostly because I feel that the entropy of life and how all these moments led to here, gives me a large entertainment value - maybe it will to you too!
The ride switch takes shape.
It’s 2018, and I’m in second year of college. Throughout the year I became close with this Erasmus student (who later just ended up switching schools to my one) from Germany. She was really big into startups. Me? I just barely had any idea at that point. But she did, and after awhile she started telling me what the startup scene in mainland Europe was like. While she and I definitely argued a lot about random things - we did have a solid friendship building along the way. It was almost like we enjoyed debating random topics and being stubborn about our opinions, classically.
After drama club one day; she told me about this other club: Enactus that we had at our university. I didn’t really like these sort of things given that my experience up to that point with group projects of any kind were:
- Join a team/group.
- Realize I didn’t agree with aspects of <thing>.
- Try to find middle ground with person in charge of team.
- Get socially excluded/ignored. Then leave to do my own thing.
So I was actually a little bit surprised that people there were willing to try compromise or take the time to explain why <idea/opinion> wasn’t exactly feasible or considered some <consequence>.
While I probably was pretty naive, and misinformed about a lot of things, not the least my opinions of how <things> should be done in an early startup. I realized at this point that this was just barely scratching the surface of what a properly executed startup should look like. Something I’d only begin to fully start appreciating two years later.
Orchard Thieves in the Library.
A year later in 2019, two major things happened. Let’s run through them real quick.
I got medicated for ADD, and I discovered the joy of building software <things>, and then pitching/presenting it to an audience.
ADHD/ADD medication in tech circles has a certain stigma to it. You’re either in the overrepresentative population of people in tech who have it, or you just want to focus and build more. Either way, for me the difference between Marcus before 2019, and after is a hill I can safely die on. It gave me so many pushes to step into things that terrified me, and gave me the emotional padding to deal with rejection and uncertainty. Two agonizing elements that would be a recurring theme going forward. It’s worth a whole post in-of-itself, but not for now.
Now about that ‘discovering joy’ thingy. That exchange student (“Chloe”, not her real name, but I don’t want to identify them since I haven’t seen them in years at this point) came up to me one day and said:
“Hey. There’s this Red Bull sponsored university entrepreneurship competition my friend told me about. We should enter into it.”
So every part of me was like: “Yes, I should do this”, but equally years of shyness and anxiety were kinda worried about what I’d actually have to do when all was said and done. But obviously, I chose to listen to the former, though, i’m glad I wasn’t told ahead of time about what we’d have to do if we won the national heat of the competition.
So Chloe booked some time in our recently refurbished libraries group study offices. In her bag, a six-pack of Orchard Thieves, she laid on the table - you see, earlier on in the day she told us that we needed to record a video to enter the competition. Whether she was doing it to silence either one of our anxieties, or for the fun of ticking off drinking-in-the-library from some bucket list - it was hilarious all the same. It took about 6-7 takes, before we were happy and we just looked it once and sent it.
Canvasing for Votes.
So we were shortlisted at this stage by the judges themselves, and it was put to a public vote. Our project in short was a way for students to find more free time to hangout with their housemates, their circles and whoever really - where we’d just consume their academic timetables and private calendars, and then find common slices of availability and then just decide for them when they’re hanging out.
We messaged everyone we knew. This was a strange feeling for me because I felt sorta odd/conscious of how people would perceive me, someone who they up to that point just saw me as maybe a super quiet guy.
Chloe herself, was probably ahead of me on that emotional curve and felt no shame messaging probably thrice the people I did. Even more so, she started asking people on popular-dating-app™️.
Time passes. Basically neck-on-neck with whoever else was in the competition at that time. But one night, as we’re in the Stables, we get told over email that we’ve won. Here’s your tickets, come thru to Toronto. Lol what.
Toronto.
So at this point i’ve been to North America just once, when I was 8 for Disney in Orlando. Truthfully, I really didn’t know what it would be like. I remember the first day going to a Chick-fil-A and ordering (no clue why I ordered this of all things for the first time) a chicken biscuit and hated it immensely.
Anyway, Toronto was nice. But it’s not being here that really matters, it’s what happened here. Red Bull’s whole show here was basically my first pitch competition hidden beneath the branding. Yes it was full of workshops about taking risks (and a boatload of X-Games speakers), but it was genuinely a nice way to be thrown at “hey we’re gonna make you pitch in front of an audience we assembled from UoT about something you’ve probably only half baked at this stage.”
I learned that my coping mechanism for anxiety around public speaking was to accelerate the level of that associated stress by compounding thoughts of “fuck it, let’s do it, fuck it fuck it fuck it”, and a lot of caffeine. Like 4-5 Red Bulls worth. So I did just that, waited in the audience for our turn to pitch, and got up on stage.
I was perfecting the smallest things. Ensuring the auto lock on the demo iPhone was off, resetting the test demo and restarting it over and over to check if it works. Forcing myself to talk slower, enunciate every word properly - and to actually give myself a chance to not mess up in a setting like this.
We landed in the top ten, and we were asked to do it again two hours later. Fuck it we ball I’m thinking I felt in hindsight. And so we did, but we didn’t win. I honestly think not winning here was the reason why I actually made to the US a couple of years later: the idea we were working on was rather tired and bland - and it probably would have sucked more time from me on the guise of “being the winners” that it ever deserved. But what I kept was the feeling that it felt fucking great to build something, present it, and have someone use it. That single dopaminergic cycle became my driving force for the next couple of years, even until now.
Intermission 1: I got bored at my placement so I downloaded Nintendo DS roms.
This is a slight intermission for a funny story that happened in between the prior and next chapter.
Im at my placement at GECAS, I learned VS Basic and automated all my day to day work, that honestly was fun at the start; but then it just turned my placement into a “Run Script at 1pm, 3pm, 5:30pm”. I got insatiably bored, so I brought an old DS to work, and realized that company policy blocked USB mass device storage.
So I do what anyone would at this point, obviously: Boot a Ubuntu VM and bypass all the windows device policies. But hey I gotta hand it to the security team because the head of Security asked me to his office (they had pretty good alerting it do be seemin’) and was basically (save for him finding out about ConnectUL/the project above) going to have asked for my termination for breaking company policy.
I think during my chat with him, that being blunt about what I was doing and not hiding it was either extremely brazen, or just the right kind of honesty that defused any hostility.
Soul Killer.
Time passes, I’m still in school at this point - roughly starting my final year of study I’m still clutching onto our timetable idea in passing. Chloe got jaded, and probably annoyed at me through her projection of her sense of loss (she took losing really negatively, among other qualities I realized that didn’t make them a positive addition on ones life.) I don’t see her again at this point.
I really think of her as a random astroid coming from left field, chaotically altering my ballistic trajectory. That action was neither implicitly positive or negative, but I do believe the delta she created in my life was enough for me to thank the action, rather than thank her reasons for doing so - which I still don’t quite have a fix upon.
Time passes, I struggle with my understanding of what it means to actually have a proper project rather than the quasi-science-fair project like the one I empirically seemed to have been left with. And yet, I was having real fun. I was actively learning, actively excited for the possibility that someone would see what I was working on, and feel interested enough to spend time to find the utility in what I was offering to them. Later I would learn this was analogous to finding your first users.
In the prior few weeks/months before Covid started, I was actively getting random people at house parties to sign into the platform and to try spark it off. I remember pretty distinctly opening a Captain Morgan for someone, watch them use the platform and to see a sorta spark happen when they saw their entire timetable import itself and figure out what’s a good time to meet (after we added each other as friends.)
Some old photos of the app. "ConnectUL"
When Covid hit, it became pretty obvious that something like this would be temporarily (on the magnitude of years in hindsight) not be feasible. So I had to stop. It felt pretty awful to be honest. And while in actuality it was probably the best outcome that I stopped working on this, it felt like a huge blow after going through the whole Red Bull experience. But like I mentioned, not winning that competition allowed me to not have an ego-drain by letting go.
And so there I was. In my childhood bedroom after being kicked out of our student housing because of Covid. Wondering what the literal fuck just happened. Knowing that I had put most of my time and effort into this, and neglecting my studies (there is a video of me admitting on stage from that competition that I didn’t go to lectures for a whole semester, this was true and my parents ripped-me-a-new-one when they found out) - it felt really disheartening and utterly scary that I was going to graduate in a year (or less) knowing that most people already had their law internships/placements done.
Redux Me.
A few weeks go by, it’s Lockdown szn, 2020. Before I moved to college, I bought this canvas wall art that was hanging on my wall for the past while. Looking up at it every day from my desk, was one of the reliefs I got while being stuck in a room for many, many months. Kinda pointless art, but the colors were really nice in a way that helped create some difference from looking at the very basic pale, cream colors surrounding the parents home.
The canvas wall art.
Then came the killing of George Floyd. What struck me the most was that the level of non-constructive hateful dialog on instagram (and by hateful, I don’t mean political, or left/right wing dialogue, I mean actual racist hate) was quite insane, and that let me to think about how smaller platforms/new platforms were handling the dialogue on their own platforms. If Instagram was sorta struggling; how were anyone else surviving?
Well there weren’t. Later on I found that our early customers just were doing human moderation from a team of 4-6 people that were busy building the frontend/backend, and had other stuff to be doing. I had never touched ML before, and honestly I was wayyy in over my head. The closest I had to any sort of training/modeling data was from a semester of Applied Economics, where I was doing really basic linear regression on economics trends in RStudio.
I spent another two weeks understanding the problem some more; for me it was a insanely large cliff edge of ideas to pickup. Loss, overfitting, data arrangement, tokens, what the difference in practical terms there was between BERT and BERT Large. Not surprisingly, my biggest enemy was overfitting, the specific domain of content moderation is still this fleeting idea of binary outcomes applied to different human vernacular profiles, from AAVE to more domain-specific nomenclature from fringe groups. And so you have problems like when a soft N word was appropriate (and whether it was even reasonable for someone not even from the US to understand the cultural nuances related to ideas like that) - to how to identify self-reflective speech about ones identity in a constructive way: we <race> needs to be <idea> vs <counter idea> versus someone saying something like: that <race> in my opinion is <unconstructive dialogue>. How could you compile a dataset to represent those strata of language profiles?
The answer (not to drown the out the point of this piece) in short, was straight forward in reality - but only at the moment it became clear. I needed to find the farthest examples from the norm about hate, a self-labeled source of people self-identifying what they were saying was directly against a particular group of people. So again, I did what anyone else would (okay that joke is getting old) I scraped 4chan’s /pol/ board everyday for 4-6 months. At the end, I had 56 million rows of labeled data that I steadily labeled with open source models, and hand crafted a larger chunk of them with OpenRefine. What was interesting here was that it was collected around Donald Trump’s first election cycle, giving me an interesting set of samples to chose from.
Tinkerin’ some more n’stuff.
At this point I had the data, and I was super confident that this was high quality. I combined it with a mixture of Sherlock Holmes’ work, and a bunch of Reddit’s I found that gave me the most general arrangement of subject areas. Anything that wasn’t flagged by the open source models above from /pol/ I labeled as FringePolitical. This turned out to be a good way of reducing false positives on political speech of a right-wing demeanor; that while many can disagree - is still speech that has a place in time within discussions. Though after QAnon, it was questionable for awhile.
What I kept running into from a high-level point of view was that we were entrusting one model to be able to decide on many different labels from one single inference. How could we lesson the weight of responsibility from one model’s decision? I figured out something that I would later learn was called the ensemble method, but to me it just seemed relatively obvious - though tricky to pull off. Two models were trained on different, random-as-possible subsets of the original 56 million rows, and both asked to provide inference on the same inputs. I thought of it as simply being able to ask two separate questions at once
- Are we talking about race, religion, sexuality, or otherwise identity?
- And if so, does that specific language match extremists profiles, while not restricting legitimate usage of those ideas?
I have no idea where it came from; but I was just on my bed drawing some lines and geometry on an A4 page trying to find some idea of how to represent these two questions in a way that could create a binary outcome. In short (and because I lack the words (and rather the formal education) to describe the problem from my implicit understanding into more specific and accurate terms that would make sense,) the closest approximation was by casting the decimal values of the inference (3-5 values per model, so 8-ish intervals) to a chiral[2] trigonometry problem (I’ll attach a drawing later to help represent this better), we could represent the (very vaguely, but entirely different) latter 2-3 stages of a trapezoidal profile in so far that once we crossed a certain threshold on either of the two problems drawn on the same axis - their contribution to the outcome became stronger. In a way we were trying to represent a hair-trigger requirement with values that represented the above two questions.
At the end, a final decimal from between 0.0 and 4.0 was created. Empirically, a value between 0.7-1.0 represented the grey area that we could use more hardcoded deterministic if/else statements to tilt out of here. A value about 1 was solid, while a value below 0.7 was for removal.
We launched the above for two customers, one of which, paid us $4,000/month for around 7 months until they couldn’t afford it anymore. I think that while I was not formally educated, that I found some confidence that this specific approach had merit from consistent syncs with the customer and regularly reviewing the BigQuery data about these inferences. Equally, I took some confidence that this customer had in their words talked to everyone else before us, and kept paying even while they were about to raise bridge financing. I did switch out the models regularly (every other month), so I was confident it wasn’t just a quirk of the models themselves together in particular.
But hey, this was also my first time using an A100 back when they were super-super new, so S/O to Google Cloud for that.
Coffee Interrogation.
Two funny things I got immediately exposed to when I approached my first VC during Covid, who actually became my first investor a few weeks later. Zoom calls became the norm for signing hundreds of thousands of dollars away, and that noise cancelling microphones were dog still.
So I get on a call with the VC, and bona-fide - I have no idea what the hell to say other than it would be cool to see what they think of the idea. Which in hindsight is probably a really really horrible way of validating your idea. While most VCs are knowledgable, like you - they don’t actually know if it will be a hit or not - they make investments in this vein equally, so asking for their opinions on your idea isn’t amazing, but it’s still good to get a collective sense of opinions. Dude in a coffee shop, roasting the idea and market in a accusatory tone - but later apologizing, was different of an experience to say the least. Ireland felt like very much a “celebrate all things equally, be critical but only indirectly” growing up. Though I’m sure anyone from home reading this might have other opinions.
Time goes by, they invest - I realize that my co-founder I wanted to bring on that really had no material impact wanted 40% even though they were going to do their JP Morgan internship the following summer (expecting me to trust that they would bail on that) taught me a great lesson in learning to say no. Later on (quite literally as I write this post), I see that a package for them arrives at the coliving I’m moving out of in SF (three years later) - fucking chaotic man.
Time continues to go by, and two months after they invest I learn from someone in the know that the fund itself was going under from failing to be awarded the right to run the accelerator (and was going to give their shares of the company to its stakeholders (some of which were (in irish terms) the literal local state governments themselves.)) So lesson two: it’s okay to give the money back, and to not ruin your cap table. I did that, respectfully, and closed down the company.
Can’t catch a break it seemed like.
Entrepreneur First.
One good thing did come from that experience. I met an alumni from Entrepreneur First, Greg, on the original accelerator. I think he knew that the same fate above was going to happen, and so he probably wanted to give something back, or something to that nature. Either way, another random point of entropy that created a better outcome. Another astroid coming from left field.
I agree, and get on a call screen. What I remember so funnily enough here was that the dude was eating some hot wings or something to that effect, and got it in his eye 20 minutes into the call. To which he spent 10 minutes in the bathroom. I like to think my contribution to the problem in that moment was a suggestion to clean the keyboard to stop hot sauce cross-contamination from creating another bathroom journey. That aside, we did have a good call, and I was passed onto the next stage.
I could see the embers of another escape hatch from my bedroom. It had to be mine. I take one group interview/presentation, and three interviews following that to get onto the program. It happens, I score really well in their internal rubric - and I’m on the program. It felt good to be given another chance to grow out of the mindset of where I was currently, and to hope for more.
Two months pass; and I find my cofounder within the program. A month or so before the cutoff date for the investment committee. It wasn’t a perfect fit, but along with other things in life - it never really is. He was excited for the tech I brought from my learnings at the last opportunity. I bring that forward and upgrade it heavily, it becomes even stronger (we had early access to GPT-3 fine tuning, which we replaced one of the two models with) and we gain another two customers. We never really had many customers throughout, but the ones we did - paid well.
But alike to the general trend of life; he wanted to work on something else - maybe he thought I wasn’t strong enough to pass IC with him, maybe he just wasn’t feeling the product anymore, maybe he wanted to do something else. It was immaterial. So I was practically left to pitch to the investment committee by myself, which was supremely odd because the entire program was designed to pair cofounders. I’m grateful they extended me the trust, and opportunity (along with a few others in the cohort that year) to pitch to them solo without a cofounder.
The juncture that mattered.
It’s a cloudy day with a good amount of sun in Ireland. It’s 4pm. Every day since the original red bull competition effectively weighed in to make this happen. Every moment helped me hold myself in the hour that followed.
I get on the call. A panel of six. It was my first time meeting most of them, since I was still at home in Ireland during the program. They ask me how I was doing. And the response, although incredibly bland in hindsight, was the most confident I had ever felt up to that point, about why I was here, and what I was doing - ever:
Well it's a sunny day, and I'm really excited to tell you what I've been working on for the past few months.
I pitch to them for an hour about why content moderation was needed, and what customers we had - and what industries we wanted to expand into to provide support. What was growing inside of me was a sort of cognitive autopilot, that would take over and shield me from that past sense of anxiety - I was in control of what I wanted to say without any fear that I would remember the feeling of the anxiety itself after. It was a sort of helpful disassociation, that left me with just the facts of what happened after. Or maybe, I was becoming that person instead of just thinking it was another side of me.
I receive the investment for the new company, and I continue on to try again. All i’ve ever really known is that the act of trying again has given me a sense of comfort about failing in the future, I just would simply try again. This was me manifesting that act.
Compound it.
The investment created an opportunity to grow the company. But I had doubts that I could do that from Europe still. I was learning that the European tech scene just wouldn’t cut it for the majority of cases. A point that most will still argue about, but one that I just feel is an axiom for the time being.
So I’m in a familiar sight. The accelerator is over, I have the investment. But I’m sitting on my bed. Wondering what’s next to do? How can I not repeat the same mistakes of the last? I needed to compound my current luck was my thinking at time, I needed to make it last longer and to give me more opportunities to grow the company.
It was hard being solo because you never really knew what the right call was, and you had to practice pretty brutal self-expectations of yourself since no one else was affected by your lack of progress other than you and your ideals for what this company should become. My brutal self-expectation was that I needed to be in SF within the next three weeks and just figure out what the fuck to do next. Literally, just place yourself into the blender and see the patterns arrange in front of you to actualize upon. In the meanwhile, I had to convince the US Embassy to grant me an exemption to the Covid ban, which was yet another strange experience that I went through and figured it out anyway.
I told my dad as I booked the ticket, and my mom a week before. I knew that they would be better suited to just accepting it is happening, and to just be supportive, so delaying telling them was a way to manage the in between. I’ll never really forget that feeling as my dad and I said goodbye at the security entrance, I was basically free to make my own mistakes, my own choices - and to just exist in this sense of flux in a different country, with admittedly very short notice.
The first housemates in the US.
First Thanksgiving!
The first housemates in the US I had were a special bunch; most of us basically weren’t from the US, so we kinda understood what it was like living here from that sense. We were all basically doing startups/tech adjacent. Most of us were just there for 3-4 months. I miss them a lot actually!
Wanna go to a product hunt party?
One afternoon in November, one of the housemates just dropped in the chat:
[...] heard there was this product hunt party happening today, anyone wanna go?
This was actually an incredibly special night, because just by itself it setup the next 2-3 years (even up to today) for me in terms of opportunities. It didn’t seem like that first, if at all - but two chance encounters made it the highest-alpha night I’ve had in terms of how long into the future it kept giving.
- Person A was frankly the reason why I got an interview with YC.
- Person B created the connection between me and a future investor that helped the company out from the brink of death - almost 3 months later.
Person A, and I got along decently well at the party. We kept in touch, and I asked them one day for a referral, since we had some early but positive traction. I’m glad they gave it, because it contributed to getting seen for interviews.
Person B, and Person C (the investor) both worked together at one point. And many months later after I left the US not knowing when I’d be back - Person B mentioned me to Person C, and made an incredibly beneficial connection.
Exodus 1.
I didn’t have a US visa at this point, and I was here on a WB waiver looking to secure investments for the company. And so, it came to an end. I didn’t make it through the YC interview, and didn’t convince an investor that the idea was worth investing in, and honestly - that was my fault. I frankly didn’t learn fast enough, and didn’t ask enough questions to root out the true values of what I believed at the time I was building.
I learned a lot from going home, and having my parents decide that my trip was no more than a “holiday”, and that I needed to “have a reality check from them, and to get a job!” Now what did I learn? Well I learned that Person C invested a few days after my parents had a large confrontation with me resulting in shouting and much more. I learned that a person(s) view of what is right for you, is utterly independent from the reality of the world around us. I learned that the reality in which they chose to live in, which while valid, is by no means less or more valid than the one you chose to live in. At this point, I know that they learned something too - that there were currents much larger than our realities within Ireland - and that they too (along with I), needed to accept that we were only living through what we know, and what we know is just the tip.
I returned to the US three weeks after initially leaving, and spent the next following 3 months working on gaining investors for the company with good success. Our MRR was growing. It was steady.
Exodus 2.
US Immigration officials allowed me to re-enter so soon, given I had a literal stack of evidence that provided answers for them - though they effectively told me that I needed a visa next time. And so when the this second trip ended, I was effectively tapped out. I had no more means to enter the US anymore, without potentially infractions. I count myself lucky that the official gave me an unofficial warning. At this point we are in March of 2022.
I arrive home this time, to a more easy-going environment without much questioning about what I was doing. I appreciated that they were adapting to this new world I was living in, even from afar. But one thing that continued to show cracks between us was the fact that I effectively refused to rejoin the Irish timezone. And so, I woke up at 5pm Irish Time (9am PST), and went to bed around 9am Irish Time (~Midnight PST.) A lot of tense was created by this act, they wanted to see me; but sort of had an odd way of expressing that through tension.
I did this for two reasons:
- I was almost immaturely resistant to giving up on my dream to be in the US full time working on this. I refused to give myself any bit of comfort, and to make it easy to settle for less back home as I put it.
- I needed to be able to take meetings at times that were appropriate for customers/investors on the west coast. Equally, it kept me in sync with what was happening over there.
I spent nine months, microwaving meals that I’d reheat at 5am Irish Time. And then code into my night. A lot of questions I get from people were along the lines of “was it sustainable, truly?”. And honestly, yes it was.
- I got good sleep.
- I ate well (the microwaved meals were pre-prepared, lean, whole-grain rice/pasta + protein.)
- I got exercise (not much, but an hour walk a day at sunrise.)
- I met friends at interesting times (for me, not them.)
- For example, for a handful of Friday’s i’d go out with them clubbing, two hours after I woke up!
Fundraising was easier than navigating the US Immigration system.
The reason why I was sleeping in such odd ways, was that I was all the while attempting to sort out my immigration status, and to navigating the insane variance in the quality of US Immigration Attorneys.
When all was said and done many, MANY, months later - I had gone through three attorneys, all but the last paid utterly disgusting fees for the horrible quality and care they put into it. My first one was indifferent, just telling you how to fill out the forms while they just stamped it and submitted it for you without any effort on their part. The second handed my case off to a placement student and they fucked it badly (so much so that they offered to do the case again for free, but I declined.) And finally, and for their privacy they won’t be named - the best damn immigration lawyer i’ve seen to date.
He took the previous two hunks of twisted metal, and just turned into a winning case. It was crazy. He took the same material that got my O-1 initially refused, to just turn it right around into success (which he later did again for me for my O-1 transfer!)
But with all the hoopla about illegal immigration, it does create a weird position for me. I am both a cousin of the subject matter, but I also after feeling the pain of the system for so many months - I have a more supportive view on protecting the system. Either make everyone else do it and clamp down on illegal immigration, or make the system more deterministic and fair for all. Some may argue that it is paramount to “pulling up the ladder” after you get on the boat. But I just see it more like that my effort was a result of my determination to be here, and it can be personally disheartening to see that it was not respected. But I am not the system, so I really can’t pretend that the system should care about my feelings - that’s a bit absurd.
I relocate to the US in April of 2023, employed by the company.
We’re starting to catch up to real time.
The details between April and November of 2023 can be assumed as really the following:
- The company wasn’t doing great.
- We tried to pivot a total of three times.
- It didn’t work out.
- Investors were adding pressure to do better. They wanted updates. I fell into a hole, I gave them up until I felt like there was no clear path out of the mess. I let as many as I could know (often in person), before I had to focus on my own well being and personal circumstances.
- By November 2023 We were going to run out of money in January 2024. The remaining funds were set aside for taxes, and corporation cleanup costs.
This was my first real VC Funded startup. Everything before it was frankly hobbyist projects - but as you can tell, they created sandboxes for me to learn within. I failed really badly here. I’m sorry if you believed in me with an investment, and that I couldn’t turn it into a reality. I’m sorry that I didn’t figure out sooner what I needed to do.
I started looking for other jobs in October, getting up at 9am, doing leetcode till 5pm (not eating anything until then, it ruined my focus.) I lost hella weight. Went out on dates purely to double over for social interaction / emotional release. Someone might look at this like I’m saying this was difficult; but rather it was fun.
I’ve learned that being burnt out was a symptom not of fatigue, but of a lack of clear progress to any sense of way forward. So having a vague sense that things can, and might be better through this grinding period, was the most fun I’ve had in a couple of months. It reminded me that I still can actualize on my own willpower.
In our high entropy world, things will and do change. And that’s enough for me to keep wanting to try again.
Soft Landing
I guess I should count myself lucky. At this point (October-November), I was two months off being forced to return back home to Ireland. I wouldn’t have had any capital left to pay a salary, a requirement of my O-1 visa. So all the past years before it would have been for naught, and I’d have to start the gauntlet again.
Through a past roommate, I got a lucky break at Replit. But it was by no means an easy win by that fact alone. I was an unknown quantity to my hiring manager, to him - I knew he had good feelings, but I can also appreciate that I had no formal education, no prior jobs in the space - nothing. So the interview panel I got was quite intense. One phone screen, one technical screen, three technical interviews on the same day (that I did ask for!), and then one follow up the day after for more signal. At the end of it; they were fast to give the offer - something that gave me immense relief.
Intermission 2: What the fuck just happened for the last two years.
I mentioned that I decided to write this insanely long recollection because frankly, I was amazed at how the mosaic of these events produced this exact stateful world I’m now living within. Exercises like this remind me that there is no merit, no benefit, no sane-reason to want for things to be different; because wishing for that would create a fundamentally different outcome. Life wouldn’t be interesting, if it went too optimally - nor would it create any sense of character. I enjoyed the pain, I enjoyed the sense that my efforts would be for naught if I didn’t give it my all when it mattered. I still fucking hate that I am a passanger to it all thou.
I feel it’s almost cliche to acknowledge that others have had it worse than me. But I think that it’s sorta always implicit to any reader, that everyone’s pain is their own mountain - and while there is objective worse outcomes - this was painful to me.
The grass was actually greener, at least to me.
After I got a job with Replit, I started to focus on the idea that I was starting to become more ingrained with American culture, its lifestyle, its problems, and everything inbetween. In short, I wanted to start moving towards what my identity was becoming. My Irish-ness is my virtue, but my daily thought process is more aligned to someone who lives in the bay area. I no longer follow what happens at home, or europe for that matter - but I still care to be informed about the problems of the time before I left, though I feel that I no longer can claim to have first hand insights onto it much anymore.
I started applying for a green card, on the EB-2 category, with a National Interest Waiver on the basis of my current O-1 Visa. It was a tough feeling: that this was penultimate step to re-establishing my life - my identity. I ultimately gave up any sort of development of my previous life back in Ireland, so failing here now would result in a serious blow - and a non-negligible of my 20’s lost to time. And frankly, I don’t think I could ever feel satisified with that outcome. That idea scares me because it would desperately have begged me to become ignorent to what life I had discovered here, and to simply memory-hole it all away.
I got my green card approved a month later. But now I must wait a year to obtain it. I’m not currently at the time of writing, completely safe on an immigration basis - but I do like to remember that I got all this done by the age of 24. And that’s a good pace for me. I must survive for a little time longer (in the grand scheme of things.)
[May 12th 2024]
Entropy wants to mess up your shit Marcus. Confirm [y/n]: y
Five months into this new job. We have layoffs at Replit (yay!) - and that did not make me a happy camper. The general consensus from colleagues I’ve spoken to after the fact, was of confusion (in particular in my case - as to why me.) But there must have been a feeling that my team needed to be downsized. Still, I did “prolific” work at Replit, as the words used in my performance reviews - and still suffered this fate, so yes you can call me a bit confused, even genuinely angry.
Ten minutes after I got fired, I just went straight to twitter and asked people to share me to any of their own open roles. It’s a weird feeling that your first ever viral tweet was of misery, but I digress. But it created a lot of good will that I never really got to be a beneficiary of, before.
What I absolutely did hate was that a large chunk of the inbound was for Full Stack roles. I can do Full Stack, I have done Full Stack - but I honestly feel it to be sorta aimless of a role. It just sorta takes any soul out of what you’re building - you own everything, do everything. This is a given in any startup (of which I have great understanding of, and experience of, don’t get me wrong), but ownership of what you’re good at gives me emotional investment. That and having to do a 4 hour interview for both stacks made me want to commit un-alive - so I just declined them.
But maybe it was more to the fact I just didn’t like these companies enough to have a big enough emotional investment for a product-wide ownership, like we fostered at Replit to a degree - instead of the “do everything aspect”. Maybe I just need to start another startup and feel ownership of the goal to foster the emotional agency to own the entire stack, product, and users.
Replit was fun. Replit gave me agency, it gave me comfort when I was in my low periods (the fact that I can just contribute to something greater than me - pulled me out of some moments along those five months) - but generally I feel sour now to it all. I feel it’s a normal human emotional reaction to have, so I’m not going to try trick myself out of it until I feel I want that. Maybe my appreciation of these five months may improve, maybe I’ll forget all about them. Time will tell.
[May 22nd 2024]
Ngl, they almost had us in the first half.
After the fallout of Replit, I had 60 days to find another job - or be ejected from the country (this seems to be a common theme amongst my stories thus far.)
So after I was laid off on a particular Thursday, I ended up managing to take 22 interviews in the week that immediately followed. The idea of having to do further interviews for each of those 22 just made me throw up a lil, so I had to cut it down to five: Anthropic, Browserbase, and others.
Anthropic went well, until it didn’t. I managed to get past 3 interviews - but then was rejected. It was particularly rough having a technical interview at 8am pacific - but I learned a lot, and honestly I should have pushed back on them to make more reasonable accommodations.
I landed at Browserbase, am I excited for what we’ll do here. It’s a cool idea, at a pretty decent time for it to have come to life.
[Jun 13th 2024]
Playing with fire.
After I accepted the job at browserbase. I spent a week getting to know the team. I didn’t feel I was a great match for the team, and decided to part ways. I’m really grateful to Paul and the team for their warmth, and I’m glad we parted on amicable ways.
I decided to take a job at, of all places, my housemates company - Bland. The team is young, their traction is fierce, and they need a guiding handing on the engineering to see them to success. I often look back at times, where I had wished for a moment where I could be at the right stage of a company, and the exact right skills to transfer onto them: complete success.
I have currently immigration issues, as in - I may have messed up on day 50 out of 60 (of the grace period) - but I have faith in my persistence to remain in the US. With, or without some hiccups - I made a promise to myself to die in America one day 80 years from now. I will keep it. I will honor it.
[Jul 4th 2024]
At this point, I’m writing from today. We’ve caught up. My intention is to treat this as a continuation of the living journey, changing the tense.
A new start. 6 months on.
It’s been awhile since I’ve written. I’ve also noticed that maybe it’s a good idea to start adding the dates to these individual subsections.
The good news: I recovered from that immigration misstep. My lawyer went the extra mile, and for that I sincerely owe much legitimate debt and gratitude towards. I’m now just 4-7 months out from having a green-card based EAD - and that would be an extreme accomplishment. As a sign of passing it forward, I helped another guy get onto his H visa - and introduced him to the same lawyer.
I’ve been at Bland for six months, and frankly it’s been the best thing I could have asked for. I get to build, and lead engineering in a greenfield, supporting real outcomes for our genuinely impressive customer base for such an young company. I didn’t realize how much hype built up, where random Hinge dates, not in tech - would comment that they heard about the company.
Working at Bland has also shown me something that made me feel genuinely happy for my self-development; people verbally expressed me as stable, reliable, engaged, and conscientious. As someone who had their own troubles with neurodivergence and ADD - I feel proud of elevating myself to the table.
I’m looking back on what the concept of my life was prior to San Francisco. I benefit from the nature in which one experiences memories as fading, and I’m glad. Looking back has not served as any sort of guide for the future. I made the mistake once, of showing someone from home what San Francisco was like - and they weaponized it to drag me back into their plane of reality, back in Limerick. Never again, will I allow such a thing to happen. I’ve learned that trying to share the good in your life with those who do not wish the same for their own lives - is a fool’s errand.
One caveat to this: my parents. I hosted them for two weeks during the summer. It was interesting, two parties realizing that the persons they knew - felt like a first impression all over again. It was tense for the first few days, but of all the years of our back-and-forth, there finally came an exchange of words on the final night that healed our differences. We no longer were strangers to one another, and maybe in the future we could be closer again.
Now I’m looking forwards into my future. Who will I meet, who will I find as a partner, what - and where will I build? These questions are open. I am 25, and I have a burning passion for my future. I have tuned my risk tolerance to account for prioritizing stability for the next few months in line with waiting for my GC, but I do understand that once this is set - I am set - and I can continue to make the risks I took before this current chapter.
[Oct 27th 2024]
Perception Creates Reality
The Atmosphere roof, a happy place to process things.
It’s nearing the end of 2024 for me, and everyone else for that matter. It’s been a colourful year, a frank understatement. I lack an understanding of what it really all summed up to be - is that okay? Should I attempt to try condense it, or does that analysis destroy the perception of what it “was”.
I’ll try.
I entered this year with a renewed sense of optimism. I finally achieved some semblance of, well, some dream to “be in tech” in San Francisco - having just failed at building a startup before in 2023. And yet; I lost a real opportunity to be closer emotionally with someone special to me, got laid off from Replit, missed the jump at some career defining interviews, and received intense emotional weathering from the entire experience combined.
Though still, I managed to find another career defining position at Bland - and a space to creatively express myself as I had once did at Chatsight.
Often the cliche here would be in the stark vein of “what doesn’t kill you, makes you stronger”, even more cliche is the direct retort to such a statement - “the damage still remains.” I don’t consider myself to be weaker, or less capable after this difficult year. Though I do more acutely feel how one feels, after coming home from a brisk cliff walk along the Irish coast - with gale-force winds leaving you longing for the warmth again. I’m trying to find that warmth again, wherever it’s been lost to. Adjacently, I’m not really sure when it got lost.
I think what I’m trying to say is that I’m aware of something that changed in my perception of how I used to view my situation and others. It doesn’t feel (as) warm, it isn’t necessarily a negative feeling, but it isn’t familiar to me. If I was to properly articulate it, it would be something like this;
I stand on an ocean beach. It’s 7pm. The clouds are too dense above to provide a color gradient for a sunset. The light is low. I feel the main feature of the scene, a cold ocean breeze push against me. Waves crash in front of me. I am alone, I think of nothing in particular. I feel everything in particular. I feel a peace that isn’t a release, but an acceptance of things.
If there was a term, I would call it: Cold Healing.
I did a lot this year. I met a lot of people this year. I saw a lot of new things. I bore myself with writing such a plain statement - though I agree with it. I feel a fear of a quieter 2025, combined with a fear such another tiring year approaches. This tirade of fear, confuses me - I’m not really sure what I fear out of programmed response, or what I consciously fear. I know I fear mediocrity of myself, in light with just how much I pushed myself before. I know I can handle whatever comes - I just long for the warmth to return.
My perception of reality is one that uses the vehicle of journaling, moments from this recount, and a rekindling of desire - to understand where I want to end up. Internally, I represent this as three pillars. Where I want to be, who I want to be with, what I want to do. These pillars guided me throughout the last two years, and while the exact answers to those questions are known to me, they form an inner dream that I protect from the outside world.
Throughout the last few months of the year, I noticed less attention was heeded to these pillars. I think in a way this was the cause of why I feel less connection to my original perception of things. But equally, the curious why, from above - could have also equally caused a forgetting of this spirit. I tend to believe the former, combined with a fatigue from attempting to achieve all three at once. One really builds on top of the other in the order I presented, and often I choose to reproduce the energy I originally came to the US with, in the pursuit of these pillars - maybe this was a lot to need to process.
In 2025, I find myself without any special reason, to want to believe again in these pillars - and to hold them a bit more closer to my sphere of thinking once again. Non-specifically;
- I know I want to remain in the US. Now, and in the future.
- I know that the type of person I want to be with, was one of the same ilk I lost the opportunity to be closer to this year. I failed to appreciate the value of this person, when they were here.
- I know that my career is in tech, and one day - another attempt at my company.
This was my first full year in the US, from January to December. I’m grateful. I will find the energy to go again, and again - until these pillars are met. I will find that warmth once more.
[Dec 26th 2024]
[===============================]
ʕノ•ᴥ•ʔノ ︵ ┻━┻
[Last updated Dec 26th 2024]
┬─┬ ノʕ•ᴥ•ノʔ
[===============================]
Footnotes
[1] I use the term ‘law school’ as a cultural translation to US readers. In Ireland, it’s actually a 4-year undergradute degree, where you receive an LLB. This is followed by a few exams (FE-1) to be qualified to practice.
[2] Yes this is the best word choice of what it represents. No there is no other suitable word without requiring more than one word. Yes I know this may sound like technobable in that sentence. I promise you I considered this before putting it into that sentence. I still cringe about the fact that it was the only word that suits, and I apologize that my lexicon’s don’t have anything to replace it with.