We have difficult scaling challenges. Efficient storage and querying of high cardinality metric timeseries data, real time, practical data science like anomaly detection (https://www.datadoghq.com/blog/introducing-anomaly-detection...), application performance monitoring for multiple languages and platforms, and now logs too, etc...
We also interesting problems on the frontend. We use webgl to efficiently render thousands of hosts on the hostmap for example.
If you're looking for work feel free to reach out - caleb@doxsey.net.