Personal Work
Because I am an hands on designer and have a background in coding, I always like to devele into person projects. Below are some of ym personal projects I've done. These are meant to help keep my broad set of skills up to date and because I love learning new technologies and building thigns.
n8n
Applications used
n8n is an open-source, self-hostable workflow automation tool that connects applications and services to streamline repetitive tasks. With the rise of AI and agentic workflows, I experimented with building an end-to-end automation for AI-generated video content.
The workflow I created fires off an AI-generated content idea every 2 days, puts them into a Google Sheet, feeds them into an AI video generation tool (VEO 3), and then automatically publishes the videos (via Blotato) to Instagram. This experiment demonstrated how automation can extend creative processes and reduce manual effort.
The workflow I created fires off an AI-generated content idea every 2 days, puts them into a Google Sheet, feeds them into an AI video generation tool (VEO 3), and then automatically publishes the videos (via Blotato) to Instagram.
AI workflow I created in n8n
While not super innovative, the project surfaced challenges around setup, execution, and ongoing management, from handling API quirks, setting IAM via Google Cloud, to maintaining reliability at scale.
Summary
This experiment demonstrated how automation can extend creative processes and reduce manual effort. The lessons learned highlighted both the potential and the limitations of integrating AI into automated creative workflows.
Statleet
Applications used
A Personal Project in Basketball Stat Tracking
Statleet is a personal project I created to track my son’s basketball performance. As he grew more competitive, he wanted to see his stats after each game. Existing apps focused on team tracking and did not support a single player view or goal setting for specific metrics like free throws, steals, or blocks. I set out to design an app that addressed those gaps.
From Testing to Building
The first version of Statleet was built with Adalo, an early low code platform. The interface was straightforward, but the logic behind it was complex. I defined how to track games, wins, losses, and individual metrics, and how to calculate per game and per season averages across multiple contexts.
The first version of Statleet was built with Adalo, an early low code platform. The interface was straightforward, but the logic behind it was complex. I defined how to track games, wins, losses, and individual metrics, and how to calculate per game and per season averages across multiple contexts.
From Building to Using
Once the app was functional, I prepared it for Apple’s Developer environment and released it as a private beta. I chose not to publish publicly to avoid the ongoing overhead of a production database. For my son’s games, the app performed as intended and delivered the stats he cared about.
Starting with Prototyping
I began by designing a series of mockups in Figma to explore how the interface should work. Because stat tracking happens in real time, the interface needed to be simple, fast, and intuitive. After completing the mockups, I tested them with a group of parents to gather feedback, then moved into development.
From Using to Vibe Coding
After several seasons with the beta, the rise of AI driven development caught my attention. I experimented with Lovable.dev and set up a Supabase database to see how they worked together. In about four hours of vibe coding I generated a new set of mockups and a working prototype.
The results were strong from a design standpoint, but there were clear limitations. Credits ran out during the build, and the tool did not support publishing to the App Store. It was easy to reach a promising midpoint, but difficult to complete end to end.
