The long version

Who I Am

The longer version—how I got here, what shipping taught me across startups, finance, agencies, and big tech, and the work that came before the four products.

Carrington Dennis

Engineer & Founder · Baltimore & Washington, D.C.

Taking something unclear and making it real enough to test — AI-native products for overlooked markets.

13+ Years shipping
4 Products now
8 Prior ventures
5 Companies
Currently building
  • Kill the BodyBoxing data
  • SupaDupaMac storage
  • CoffeeBadgerWorkplace tools
  • City of ConsequencesStrategy game
Technology & tools
Languages
  • TypeScript
  • JavaScript
  • Python
  • Go
  • Java
Frontend & mobile
  • React
  • React Native
  • Electron
  • Redux
  • GraphQL
Backend & data
  • Node.js
  • Ruby on Rails
  • Spring
  • PostgreSQL
  • MongoDB
Cloud & infra
  • AWS
  • GCP
  • Kubernetes
  • Docker
  • Jenkins
AI & ML
  • OpenAI API
  • NLG
  • PyTorch
  • TensorFlow
Previous work
  • InsOuts
  • Cutman
  • Baktun
  • NiteOwl
  • PeopleOps
  • ZeeMee
  • Vyse
  • Oiree

Follow the work

Experiments, essays, and what I’m testing next.

Subscribe

The pattern

I’ve been early before.

  1. Learned Unity and computer science
  2. Automated a trading desk’s written commentary, years before generative AI
  3. Building on React Native while it was still new
  4. Now AI-native products for overlooked markets: boxing data, Mac storage, private workplace tools, a strategy game
On this page

This is the longer version of the About page—who I am, how I got here, and what I've learned building and shipping software for more than a decade.

Four products, four different bets

Building four products at once may sound unfocused.

For me, it is a way to test several observations against reality.

Each product has a different customer, market, business model, and set of assumptions. Each one forces me to think differently about product design, distribution, technology, and what people are actually willing to pay for.

The goal is not to keep every product alive forever.

The goal is to learn which problems are real, which solutions people care about, and where my judgment is right or wrong.

Not every idea I pursue is completely original. That does not bother me.

Sometimes the opportunity is not inventing a new category. It is looking at an existing problem from a different angle, making a sharper set of decisions, and building the version you believe should exist.

The technology is not the point

I work across the whole stack, from the interface people use to the systems running underneath it.

AI has changed how much of that work I can take on myself. Research, design, engineering, operations, content, and analysis can now be compressed into a much smaller team—and sometimes into a single founder.

That shift gives me more room to explore ideas, move quickly, and build with a level of ambition that once required far more people and capital.

But the technology is not the point.

A product can be technically impressive and still fail to matter.

As an engineer, the temptation to begin with the build is always present. You have a cool idea, convince yourself it should exist, and start writing code. AI makes that temptation even more intoxicating because it has never been easier to turn a half-formed idea into a polished demo.

I have seen enough pitch competitions, startup showcases, and product demos to know that something can look convincing without being useful.

So I try to begin somewhere else.

I observe first.

Shipping is part of the research

I believe there is value in shipping.

Not simply because finished work is better than unfinished work, but because shipping exposes what thinking alone cannot.

A product changes once it reaches another person.

The assumptions become visible. The awkward parts become harder to ignore. People use it differently than expected, misunderstand what seemed obvious, or care deeply about something that felt secondary during the build.

Shipping moves an idea out of your head and into contact with reality.

What counts as shipping, however, changes across industries and business models.

For a consumer product, shipping may mean placing something in the hands of thousands of people and learning from attention, retention, habit, and word of mouth.

For a B2B product, shipping may begin with one customer, one team, or one narrow workflow. The software is only part of the problem. You also have to understand budgets, procurement, security reviews, organizational incentives, and how buying decisions are actually made.

In hard tech or deep tech, the feedback loops are longer and the cost of being wrong can be much higher. Shipping may mean completing a prototype, passing a technical milestone, manufacturing a component, running a field test, or proving that one difficult part of a system can work reliably.

The artifact changes. The principle does not.

Eventually, the idea has to meet the world.

That is one reason I am interested in building across different markets. Each develops a different kind of judgment.

Consumer products sharpen taste, clarity, and distribution.

B2B products force you to understand workflows, budgets, incentives, and institutions.

Hard and deep technology demand patience, technical rigor, capital discipline, and respect for the physical world.

I do not believe every company should be built the same way. I want to understand what shipping requires in each context—and what kind of builder I need to become to do it well.

I've been doing this for a while

I started building personal projects around 2010 and began working professionally as a software engineer in 2012.

I built for the web during the JavaScript and jQuery years. I moved into mobile as the modern app ecosystem was taking shape. I worked with Unity as it became a standard tool for independent game developers and later adopted React Native as cross-platform development matured.

Since then, I have worked across startups, financial institutions, agencies, consulting engagements, internal platforms, mobile apps, games, consumer products, and side projects that never became companies.

The tools changed.

The part I enjoyed did not: taking something unclear and making it real enough to test.

Over time, I also learned that every environment has its own definition of shipped.

At a startup, shipped may mean getting something useful into a customer’s hands before the company runs out of money.

At a large company, it may mean navigating dependencies, security requirements, legacy systems, and several teams before a change reaches production.

On a personal project, it may simply mean resisting the urge to keep polishing and finally letting another person use it.

The work is not finished when the code works. It is finished when the product survives contact with the people and systems it was built for.

Previous work

Before the four products I am building now, there were other products, experiments, client projects, and attempts:

InsOuts · Cutman · Baktun · NiteOwl · PeopleOps · ZeeMee · Vyse · Oiree

Some shipped. Some stalled. Some failed. Some were work for other companies. Others were ideas I could not stop thinking about until I built them.

Together, they are the better part of my education.

For each project, I want to document four things:

What it was — the problem, the intended user, and why I believed it deserved to exist.

What I contributed — what I designed, engineered, led, or helped bring to market.

What happened — whether it launched, found users, stalled, failed, changed direction, or became something else.

What it taught me — the lesson I carried into the next thing.

Not every project needs to become a successful company to have value. Sometimes the value is technical. Sometimes it is commercial. Sometimes it is learning exactly why an idea did not work.

How I got here

  1. 2013

    University of Maryland, Baltimore County

    BS, Computer Science · Minor, Game Development

    I graduated from UMBC and began turning a long-standing interest in computers into a career in software.

  2. Baktun

    Founder · early product attempt

    I started one of my earliest serious product attempts.

    It began teaching me the distance between having an idea, building a product, and creating a viable company.

  3. 2012 – 2016

    T. Rowe Price

    Software Engineer I · Solutions Analyst

    I started my professional career in finance at T. Rowe Price.

    There, I saw how software operates inside a large institution, where reliability, regulation, process, and organizational complexity matter as much as the code. I enhanced GlobalNet, the firm's internal platform, and automated trading-desk commentary—cutting reporting time by roughly 40%.

    I also worked with natural-language-generation systems years before generative AI became a mainstream subject. That gave me an early look at software that could produce useful work on behalf of people rather than simply present information to them.

    • Java
    • Spring
    • Struts
    • JavaScript
    • jQuery
    • PostgreSQL
    • Jenkins
  4. 2016 – 2018

    InsOuts

    Founder · Full-Stack Engineer

    I left T. Rowe Price to pursue InsOuts, one of my first serious startup attempts.

    InsOuts used Bluetooth beacons to help barbershops recognize when customers arrived and manage their waiting rooms—an algorithm I designed cut customer wait time by roughly 63%.

    The product did not become the company I imagined. It taught me that interesting technology, working software, and a viable business are three different things.

    • React
    • React Native
    • Objective-C
    • Node.js
    • MongoDB
    • AWS
  5. 2018 – 2019

    Fixt

    Full-Stack Engineer II

    I joined Fixt because I wanted to see how an early-stage company was built from the inside.

    I wanted to understand the work beyond engineering: customers, operations, hiring, limited resources, and the constant negotiation between what should be built and what could actually be built.

    • React
    • React Native
    • Redux
    • GraphQL
    • Ruby on Rails
    • SQL
  6. 2019 – 2021

    Independent contracting

    Contract software engineer · NiteOwl & others

    I went full time as a contractor while continuing to search for ideas and products worth pursuing—including end-to-end mobile app development for NiteOwl.

    Working across companies exposed me to different industries, teams, codebases, and definitions of good work. It taught me how quickly you have to understand an unfamiliar system before you can contribute something useful.

    • React Native
    • Node.js
    • Express
    • MongoDB
  7. 2021 – 2023

    8th Light

    Full-Stack Engineer III · software agency

    I joined an agency to become a more disciplined engineer. There, I built mobile testing infrastructure for Riot Games' Valorant—an iOS test suite and a greenfield API for test-account management.

    That period sharpened how I thought about clean code, software architecture, maintainability, communication, and the responsibility of leaving a system better than I found it.

    • .NET
    • Python
    • Appium
    • Jenkins
    • Perforce
  8. 2016 – present

    More products

    Founder · Carrington Creative

    I kept building outside of work—commercial products, experiments, and tools born from problems I could not stop thinking about, like SupaDupa AI and Unblock Art.

    Each one helped me develop a better sense of what I enjoy building, what I am good at, and where my instincts tend to be wrong.

    • TypeScript
    • React
    • Electron
    • Python
    • OpenAI API
    • GCP
    • AWS
  9. Oct 2024 – Oct 2025

    Capital One

    Technical Lead · Individual Contributor

    I joined Capital One and worked on software inside one of the largest financial institutions in the country. I led a team of five engineers building multi-factor authentication infrastructure across apps, sessions, and devices, and retired legacy Java systems to simplify the architecture.

    The experience taught me how systems behave at scale—not only technical systems, but teams, incentives, dependencies, standards, and institutions.

    • Go
    • Java
    • AWS
    • Kubernetes
    • Docker
    • Jenkins
  10. Now

    Carrington Creative Studios

    Founder & Engineer

    I left Capital One to understand how AI is changing the way products and companies are built.

    I am now building Kill the Body, SupaDupa, CoffeeBadger, and City of Consequences through Carrington Creative Studios.

    • TypeScript
    • React
    • React Native
    • Electron
    • Python
    • OpenAI API
    • AWS
    • GCP

The world gives off signals constantly.

I'm paying attention — and shipping things real enough to find out whether I was right.

Follow the work

Experiments is where I share what I'm building, testing, and learning along the way. Essays is where I work through the larger ideas behind the products, industries, and technologies that interest me.