Don’t make the mistake of thinking that you should build everything yourself.

I have made this mistake, and I continue to see it being made among investment managers.

Most of us know how to use Python, and we have usually written some code for mathematically advanced models at university.

However, the skills required to build fast, stable and secure implementations are very different from that.

Even for common problems like linear programming, the best solvers come from proprietary solutions.

Great implementations require years of obsessing over the problem and discovering various small tricks that in combination produce fast and reliable results.

If you work in a bureaucratic organization where you don’t have full control over both the software and infrastructure, the probability of succeeding with such a task is practically 0.

Save your precious time and avoid falling technologically behind.

For more on CVaR and CML, see: