ChatGPT. I did not expect to be in the top 1% of messages sent and in the first 0.1% of users.

I started running into the issue of running out of space on my iPhone and Mac, even with larger storage options, due to an ever-growing collection of text messages. Instead of paying for more iCloud storage and dealing with performance issues from having 10 years of messages, pictures, and videos on my phone, I decided to export and backup the messages. I then set the “auto-delete after 1 year” option in my iPhone settings to clear up my storage.

The process was a bit complicated, so I wrote a guide on how to do it and published it on my wiki at tehguide.com. If anyone else is facing the same issue, check it out here:

https://tehguide.com/export-iphone-text-messages

I ended up at an open mic last night. It was mostly people trying their hand at being comedians.

There was one guy who went up to do a set. I actually sat next to him when I first arrived. Mid 20s, white, great hair, so fit he looked like a bodybuilder. Apparently he was diagnosed with brain cancer (recently or not, I forget that part of the story – he didn’t seem like he was lying, before someone suggests that).

He was joking about the fertility clinic, after the doctor told him that he needed to bank some sperm if he doesn’t want his future kids to have child defects. Something associated with the cancer, or maybe the medication? His jokes were about how he was the only dude in a waiting room with lots of moms, and how they said he could have company if he needed it and how there couldn’t be any lube or saliva in the sample. It was pretty funny, even if my reprinting of it isn’t.

Seemed like a genuinely positive guy, who wasn’t going to let brain cancer have any part of his life as long as he could help it (unless it helps him in comedy of course).

I don’t really know what to make of it, honestly. Just one of those unexpected moments where you sit down next to someone, and they turn out to be a little more complicated, or impressive, or strange than you’d expect. And then they go up on stage and make a room full of people laugh about something that should be devastating. I’m still thinking about it.

6/13/25 Update: Since writing this, I have found a full-time role with which I am quite happy. That said, I am still interested in ideas, etc, so if you’ve got something in mind, please don’t hesitate to reach out to me.

—

About a year and a half ago, I left my job as an Operations Engineering Manager.

Since then, I’ve done disaster relief volunteering, run my own tech support business, worked on startup ideas, and spent time thinking about what I should do next.

A few months ago, I started seriously applying for full-time jobs again. My only requirement? That I actually believe in the company’s mission. That cut out a huge chunk of the market—no finance, no real estate, no insurance. Instead, I focused on companies doing real good: mental health services, robotics, AI for everyday people, tools that improve the world rather than extract from it.

At the same time, I want to keep working on things that feel meaningful. So I’m putting this out there: If you need something built or improved—an app, a website, a business, a service, a game, a robot—I’d love to help. I’m not looking to charge much; I just want to work on projects that actually lead to something real.

If you’ve got an idea or know someone who does, let’s talk.

Email me at roman@roman.cx, or leave a comment below.

I recently came across a guest essay in the New York Times titled “UnitedHealth Group C.E.O.: The Health Care System Is Flawed. Let’s Fix It.” (original article, alternative non-paywall link).

With a title like that, I was excited. Surely, this was going to have some substance—especially coming from a publication like the NYT. But I was disappointed.

The article offered no real solutions, no clear plan, and no concrete actions from the author or UnitedHealth Group itself. Instead, it was packed with platitudes about increasing transparency, providing individualized care, and building partnerships. These are nice ideas, but they’re meaningless without specifics about how to make them happen.

It made me reflect on something I’ve always found odd about how Americans view public services. Firefighters, police officers, and public education are not built around maximizing profits, yet healthcare—something so intimately tied to life and death—is treated like a business. Why is a police commissioner’s salary capped at $300-500k, while insurance CEOs make $10 million or more? Should staying alive and healthy be considered a luxury?

The deeper issue here is something the article tiptoed around: the health insurance system is fundamentally built on profit incentives, and that often pits the goals of insurers against the needs of patients. Insurance companies make more money by collecting premiums, denying claims, and restricting care. As long as the system has significant rewards for that kind of behavior, it’s hard to imagine meaningful change.

To be fair, profit incentives have driven real innovation in parts of the healthcare industry. In pharmaceuticals and medical technology, for example, companies have developed groundbreaking treatments and devices because they were motivated by the potential for high returns. Even in healthcare delivery, innovations like telemedicine and urgent care models have improved access and convenience.

When it comes to health insurance, the profit motive seems to do more harm than good. The focus isn’t on improving outcomes but on managing risk and cutting costs—often at the expense of patients. This is why we see practices like claim denials, complex coverage terms, and surprise billing. They might help a company’s bottom line, but they don’t serve the people who rely on health insurance to stay healthy.

The system is flawed, but it won’t change until we address the core issue: health insurance companies are incentivized to prioritize profits over patients by denying claims and restricting care. While profit can drive innovation in some parts of healthcare, as long as the system rewards putting dollars over providing the most effective care, it will be corrupted against patients.

I was recently asked to provide my opinions on the current state of the generative AI industry.

1. Who is the winning player and how deep is the moat (and what is the source)? Key differentiator?
   • The winning player is OpenAI, with 180 million active users. Google is close behind with 145 million active users. OpenAI’s ChatGPT has over 175 billion parameters, and the source is a diverse range of text data from the internet, including websites, books, articles, forum posts, and other publicly available written content. Key differentiators are being one of the first to market, one of the biggest models, and high quality of results.
2. What risks are associated with the adoption of AI model training platforms, such as technology obsolescence or vendor lock-in?
   • The risks are unpredictable ongoing costs, high setup costs, data security issues, and ethical concerns.
3. Could you elaborate on the current trends shaping the Gen AI market, particularly those that are influencing decision-making processes at the C-suite level in smaller companies? How do these trends differ from those affecting larger corporations?
   • The current trends include higher demand for personalization and increased usage across different industries. In terms of decision-making processes at the C-suite level in smaller companies, the trends are about trying to find a good cost-return balance, versus a larger corporation, which would be less concerned about costs and more about security and long-term success.
   • A significant trend that is common across different sized companies is the desire to jump on and maximize benefits from Gen AI as quickly as possible.

After 6 amazing years at Datadog, I’m moving onto new adventures.

For much of my time at Datadog, I worked with very smart, kind, and collaborative people. For whom work meant more than a line on their resume, and their coworkers are more than just people with whom you spent 8-9 hours almost every day.

I’m very grateful to have met and worked with them.

? Highlights of my journey at Datadog:

• Getting to work with the amazing Solutions Engineering team
• Fulfilling my dream of becoming a Software Engineer
• Implementing cutting edge ML solutions, and helping design, code, and launch brand new software
• Learning meaningful leadership lessons

I’m excited about my next adventures. I’m hoping to put my experience and skills to meaningful use. I’m taking a small break and looking into starting my own revenue-generating initiatives.

For anyone still working at Datadog, I’d love to stay in touch. Please feel free to connect with me here on LinkedIn or reach out to stay updated on my journey or discuss potential collaborations.

To my fellow Datadogs, thank you for your support, your camaraderie, and everything.

We have (had?) a tradition in Solutions Engineering to recite a haiku for big events like promotions or departures, so here is mine:

Six years of yaml

Also code, friends, work, projects

Forever grateful

I had a tiny wood making project over the last two weeks, involving making a more comfortable place for my cat to poop.

I got a litter box enclosure from Etsy 1-2 years ago but the size I got left my cat a bit too little vertical space. So I found a shelf in my buildings basement, ordered a jigsaw from Amazon for $35, and made four little raisers for each corner. Turned out pretty okay I think.