I'm the founding engineer at Enttor, where we use AI and browser automation to make outbound marketing actually work. Before that, I shipped an e-commerce platform at Samsam. M.Sc. in deep learning from Universidad de los Andes.
Conway's Game of Life · a zero-player cellular automaton. Each cell lives, dies, or is born from the count of its 8 neighbors (B3/S23). Simple rules, surprising patterns. Click to seed a stamp. read more →
A tighter edit of the systems behind the resume: product surfaces, automation loops, and experiments that still feel alive.
The system finds prospects on Instagram and LinkedIn, qualifies them with OpenAI pipelines, and moves the work through dashboards, APIs, and queues. I built the browser automation, the product surface, and the backend rails that keep it moving.
A WebAssembly cellular-automaton sandbox: Conway, Wireworld, Brian's Brain, Lenia. Real-time on a million cells, all running in the browser. The hero on this page uses a sibling implementation.
An interactive atlas of escape-time fractals · Mandelbrot, Burning Ship, Julia, Newton · with deep zoom, color schemes, and shareable URLs. A weekend project that ate three weekends.
Number-theory walkthroughs as live Jupyter notebooks: Ulam spiral, Goldbach pairs, the Riemann staircase. Used by a few math teachers I admire.
The page you're on. Open-source, built incrementally, and a place to put math that I think is beautiful next to software I'm proud of.
Founding engineer, full-stack, deep-learning grad. Short version below; deeper version in the work.
I'm a software engineer from Bogotá. At Enttor I built the AI outbound engine from the ground up: the browser automation that finds prospects on Instagram and LinkedIn, the OpenAI pipelines that qualify them, the Next.js dashboards where it all lives, and the NestJS APIs and Inngest queues holding it together.
Before Enttor I was founding engineer at Samsam, an e-commerce marketplace with a React Native app for shoppers, a Next.js dashboard for merchants, all on Prisma and Postgres. Earlier work spanned analytics (SQL pipelines and Power BI dashboards for HR, risk, and operations) and deep learning on satellite imagery to spot croplands and flag environmental risk.
B.Sc. in Systems and Computing and M.Sc. in Information Engineering, both from Universidad de los Andes, with a deep-learning specialization and a minor in management.
Two founding-engineer tours, a master's in deep learning, and the long stretch at Uniandes that taught me how to ship.
I built the system that finds the right prospects on Instagram and LinkedIn, qualifies them with AI, and starts real conversations, at scale. Full-stack: Next.js dashboards, NestJS APIs, Inngest queues, all running on Vercel and Supabase.
Shipped both sides of a marketplace: a React Native app for shoppers, a Next.js dashboard for merchants. Cut backend response times by 10% and built an alert system that surfaced critical errors 70% faster.
Trained deep-learning models on satellite imagery to spot croplands and flag environmental risk. Taught as a graduate TA in parallel; designed labs, ran office hours, helped underclassmen ship their first models.
Five years on the foundations: systems, algorithms, applied ML, plus a minor in management. Side projects in Python, TypeScript, Java, and Swift along the way.
A few stamps along the way.
Awarded for ranking in the top tier of the country’s national high-school exit exam (Saber 11). One of Colombia’s top student honors.
Building, deploying, and scaling cloud-native apps on AWS: IAM, Lambda, DynamoDB, S3, the rest of the toolbox.
AWS fundamentals: services, security, architecture, and the billing model that decides which of those you actually use.
The repos, the languages, and a year of real commits, straight from GitHub.
Short essays on engineering, startups, and what shipping actually feels like.
⋯
You can be a great engineer and still be poor.
Not because you can't code. Because you can't speak simply.
I love solving problems with technology. Building systems, automating repetitive work, making stuff work.
But I learned the hard…
⋯
If you vibe coded your app, it's probably already hacked and you don't know it.
AI ships code that runs. That's development. Ask AI why it picked this pattern over another and you'll get a confident answer that means nothing. Engineering…
⋯
Last week I got my master's degree.
I started my master's the same week I started my first full-time job at a startup.
Early 2024, I was going to work in the day and going to class at night. The startup was full of my friends from…
⋯
Building a startup feels a lot like preparing for a marathon
At the beginning, the excitement is overwhelming. You feel ready to take on the world. You have a vision, a plan, and the motivation to match. The first kilometers feel almost…
⋯
Since February 2024 I've been working at a startup my first job and, honestly, a turbo charged master's program. These are a few lessons that have learned:
Here it's okay to fail, as long as tomorrow you're better A merely "decent"…
⋯
One moment you're debugging with your team, sitting together at your desks. The next, you're walking out of the Apple Store with a Mac mini under your arm.
Last week, we faced a frustrating blocker. Instagram kept throwing captchas and…
Type a command: help, whoami, experience… Need a real conversation? kevin "your question" opens the LLM.
welcome · a small shell that indexes this site. type help for commands, or ask the AI directly: kevin "what does enttor do?"
This is just a corner of the internet where you can get to know me a little. If something here resonates and you want to say hello, the inbox is open.