More About Me

I spent the last twelve years as the founding engineer and technical anchor at a startup that scaled to serve tens of thousands of businesses. I originally joined as a contractor to port a legacy prototype to a modern web stack, mostly because it was impossible to hire developers for the old one. Within a year, the team shrank down to just me. From then on, my day-to-day was keeping the system alive, scaling it, and acting as the strategic bridge while my partners built a multi-million-dollar business.

Because we operated under tight constraints, I couldn't rely on standard, heavy solutions. I had to learn FinOps the hard way. I didn't just monitor dashboards; I actively stripped out inefficiencies. By aggressively implementing edge caching, I eliminated over 850 million API requests a month. By rethinking our DevOps strategy, I consolidated our Kubernetes clusters by 75% and dropped our backend P99 latency from 400ms to under 25ms in the process. When our image pipeline reached the point of processing billions of images, standard tools became too slow, so I wrote a custom byte-level stream parser based on the raw JPEG spec to extract metadata on the fly.

We also relied heavily on external vendors, which meant I spent a lot of time debugging black boxes outside our codebase. Whether that meant reverse-engineering a multi-megabyte minified JavaScript bundle to trace a WASM memory leak, or reading network protocols to pinpoint a failing third-party data center, I had to learn how to diagnose unfamiliar systems from heuristics alone. When a massive spam attack threatened our domain reputation, I couldn't just wait for a vendor to fix it. I engineered a custom honeypot and edge interceptor to neutralize it at the network layer.

Of course, scaling a platform to that size is never a solo effort. We built a great team over the years, but as we grew, my role naturally evolved toward our deepest systemic issues. I spent my final year maintaining zero downtime as a sole operator during a major team transition, ensuring the platform remained rock-solid. I am naturally analytical and strategic, but to me, that just means I care deeply about the fundamentals. That decade in the engine room taught me that there is no substitute for knowing how things actually work under the hood.

Now, I'm looking to bring that experience to a new team building something great. If you are tackling hard systems problems, wrestling with runaway infrastructure costs, or just want to talk shop, send me a message.

Recent Technical Impact

  • Performance: Reduced backend P99 latency from 400ms to <25ms, and dropped backend memory footprint from 8GB to 450MB.
  • FinOps & Infrastructure: Consolidated Kubernetes nodes by 75% (16 to 4) and removed hybrid caching layers, saving tens of thousands in annualized compute costs.
  • Scale: Dropped image resizing worker calls by 82% (from 750M to 130M per month) and eliminated 850M API requests per month via Edge Caching.
  • Security: Singlehandedly neutralized a critical spam attack by engineering a custom edge interceptor and honeypot system to protect email deliverability.