Mark Stickley

Hi, I'm Mark Stickley, a professional software engineer passionate about building beautiful software with a focus on intuative usability. I have a strong background in modern JavaScript frameworks, UI/UX best practices, and a love for clean, maintainable code. Welcome to my personal site!
This website is also an experiment: it has been built entirely using AI. I haven't written a single line of code myself—instead, I've directed my trusty AI companion (GitHub Copilot, mainly using various Claude models) to design, develop, and test every feature from scratch, based solely on my written instructions.
The experiment has been largely successful: the AI was able to generate all the required functionality and save significant development time. However, while the process was much quicker than building everything by hand, it was not instant—substantial technical direction and decision-making were still needed throughout. This approach is not yet something an inexperienced coder could undertake without a strong understanding of web development. It was also quite frustrating at times, as the AI occasionally misunderstood instructions or produced code that didn't work as intended, requiring multiple iterations to get things right. The code itself is not something I would be proud to show off, as it lacks the finesse and optimization I would typically apply myself. However I was able to sneak some complicated functionality into a relatively basic site without having to do the usual research and coding myself, which was satisfying.
(Complicated functionality that might not be immediately obvious include the client-side decryption of my private CV data, using PBKDF2 and AES-GCM via the Web Crypto API; and the dynamic generation of skill rating bars based on either numeric ratings or textual levels, and the inclusion of a link that opens the user's email client with a pre-filled subject and body while obfuscating my email address from bots.)
Overall, this experiment has shown that AI can be a powerful tool for accelerating web development, but it is not yet a complete replacement for human expertise. With further advancements in AI capabilities and better tools for guiding the development process, I believe we will see more widespread adoption of AI-assisted coding in the near future.
Addendum (February 2026): The above text was written in December 2025, reflecting experiences between August and December of that year. It's worth noting that even in the relatively short time since then, AI agentic coding capabilities have advanced significantly, making the earlier assessment feel somewhat outdated already.