Timothy Mensch

Designed the back-end infrastructure and wrote the core software to allow for high quantities of data to be collected and processed via a machine learning algorithm to detect specific patterns to help customers increase their profitability.

When I was hired, the current infrastructure was buckling under a minimal load; the company hadn't even scaled up to full capacity and it was already failing. I reworked the architecture to be such that their server costs would be low, which was critical to support their market price point, saving them thousands of dollars per month compared to the approach they first asked me to use. This cost savings enabled their business model to be profitable.

The design also relied as much as possible on managed services, so that their ongoing IT expenses were minimized.

I led the back-end and architectural development on a team that created software to automate the sales quota and allocation process of a large multinational technology company.

The project was a challenge because no one person at the company understood all aspects of the sales process. We understood this going in, and I created a very flexible design that was able to change quickly as new requirements were uncovered from the clients.

The sales process rules were also quite complex. An organic process that had evolved over the years, it was adapted well to humans but not as well to digital logic. After we worked out a consistent algorithm the computer could apply reliably, the resulting algorithm enforced three dimensions of constraints (time, sales hierarchy levels, and sales people) for every item in their catalog. The design was well tested and extensively covered by tests to ensure it behaved as they required.

The truth is that a process that seems simple to the humans who follow it may not be as simple when coded into software. Part of my expertise is in distilling such an algorithm down to its component parts and ensuring it will scale and evolve along with your project and user requirements.

I worked with a startup to design a core back-end technology stack that would support their SaaS product.

As the product has a free tier, it was imperative that the incremental cost-per-user was as low as possible. I designed and implemented the core architecture in such a way as to minimize their ongoing server costs while allowing it to be easily extended by adding additional features that would add value that some customers would be willing to pay for.

The system we created was able to run at an extremely low cost per user ($0.012/user/month), enabling the company to scale quickly and charge reasonable rates to premium users.