KYLE SHANK

Structure

Navigation and structure are important parts of a software product. They carry meaning and signal what about a product is prioritized.

Structure is especially important for web applications. URLs still have meaning in my opinion. You should be...

Launch day

Launch days are usually underwhelming. After several launch days I’ve learned that people tend to overestimate external interest. This is especially true for apps that at this point take a considerable amount of effort to persuade potential users...

Narrative drives startups

For a long time I thought building the best possible product was the most important goal for a startup. I was wrong.

The thing a startup cannot live without is a strong narrative. The narrative is what investors pay attention to. The narrative...

The Right Analogy

Sometimes when you’re building a product finding the right analogy can help answer a lot of questions. I’m building out FANS.FM and have struggled with how to position the product. What should FANS.FM look and feel like to draw out correct assumptions...

Machine learning is hard

Public Service Announcement: Machine learning is harder than you think.

I spent 2 years trying to build a consumer facing machine learning product and failed. Getting a machine to learn takes tons of data, tons of experimentation and tons of data...

Personalization is a Problem

We all live in a filter bubble. Our streams of information are curated to our likes. We tend to follow people we agree with and block those we don’t. We all maintain an illusion that we control what we consume but in reality we are greatly influenced...

Marketing driven product decisions

Silicon Valley on HBO gets a lot right about what its like to work in a startup. Last weeks episode nailed what it feels like to be an engineer that gets railroaded by marketing driven product decisions.

The thing they got so right about it was just...

Programming is Easy. Product is Hard.

The bottleneck is never what you think it is.

Development capacity is often blamed as the reason something isn’t done yet. More often than not though the reason something isn’t done yet is because you can’t decide what to build. Instead of focusing...

Early Success

The biggest thing I’ve done in my software life I did over 10 years ago. The RadRails project was fun and exciting. It was a right place, right moment, right time, right brand kind of moment that we seized upon and it worked.

Over 150,000 developers...

Logic First

The most important thing in a software system is it’s logic. What to do, when and how is the problem software solves. Unfortunately, the logic of a software system becomes less and less clear over time to the developers that work on it.

As a piece...