Bloomberg receives large volumes of high-frequency financial time series from a wide variety of sources. We strive to provide to our clients real-time, reproducible and top quality analytics derived from this data. In this talk, we will present to you the challenges associated with running analytics on top of this complex data and the solutions we have built for anomaly detection, prediction, aggregation and scoring of data. We will delve into our technology stack which comprises of many open source technologies such as Apache Spark (parallel computing engine), Apache Cassandra, Apache Zeppelin (development and visualization notebook). We will also showcase how we use Machine Learning to enhance our data quality on a daily basis.