Senior Software Engineer - News Search Relevance
Posted Jul 18, 2018 - Requisition No. 68177
News moves markets, and with an index of almost half a billion stories in dozens of languages, we face a huge task to surface the most important information to help our clients make the right decisions.
The News Search Experience team has built a system which uses machine learning to rank stories returned in millions of searches each day, in some of the Bloomberg Terminal's most used screens.
The competitive environment our clients work in mean we only have milliseconds to process each search, whether that be on a single company or a thousand-character long query covering a portfolio.
We rely heavily on open source technology including Solr/Lucene and the Hadoop stack, and we have co-authored the first machine learning based ranking platform for Solr which has been contributed back to the community.
As a Senior Software Engineer, we'll trust you to:
- Iterate quickly to develop new rankers
- Implement the full pipeline, from offline data analysis (leveraging tools such as scikit-learn) to find new features with which to train models
- Build the low latency infrastructure to extract these features in production, train new models and measure the impact of new rankers on search quality
You'll need to have:
- Experience developing low latency data pipelines in Java and/or C++
- Experience building and debugging distributed systems
- Knowledge of fundamental supervised machine learning algorithms and evaluation techniques
We'd love to see:
- Knowledge of the Hadoop stack of applications (HDFS, Spark etc.)
- Knowledge of Solr/Lucene
- Interest in open source
For more information on what we've built, please have a look at these videos:
If this sounds like you:
Apply if you think we are a good match! We will get in touch with you to let you know what the next steps are. In the meantime, check us out at http://www.techatbloomberg.com/
Bloomberg is an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, colour, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.