One of J.P. Morgan's top data scientists has left
The head of data science research has left J.P. Morgan. Insiders say he's off to join Deutsche Bank.
Named as one of our top 20 data scientists in banking and finance, Graham Giller was head of data science research at J.P. M between January 2017 and February 2018. Before that, he was the chief data science officer and a manager in the office of advanced analytics.
J.P. Morgan poached Giller from Bloomberg in 2013. At the time, his arrival was seen as something of a coup for the U.S. bank. As an experimental elementary particle physics PhD with a history of building predictive analytics and machine learning systems in finance, Giller was that rare thing - an academic with a proven history of commercial achievement.
His exit from J.P. Morgan comes after the U.S. bank lost its former head of machine learning, Geoffrey Zweig, to Facebook in January. At J.P. Morgan's recent investor day, CFO Marianne Lake stressed the bank's interest in developing machine learning tools in 2018. However, J.P.M's strategy seems to be in disarray. - After hiring Zweig as an expert in natural language processing in 2017, J.P. Morgan subsequently promoted Samik Chandarana, a former credit trader, to head its data science and analytics (effectively its machine learning) strategies in October last year.
Deutsche Bank didn't immediately respond to a request to comment on claims that it's poached Giller. The German bank has built a large "strats" team under Sam Wisnia, the former co-head of strats at Goldman Sachs, whom it hired in 2014. Giller's alleged arrival suggests Deutsche is ready to take its use of data to another level.
Meanwhile, the exodus of machine learning specialists from J.P. Morgan suggests banks might want to rethink their data hiring policies. J.P. Morgan boasted last August that it was years ahead of other banks in the implementation of machine learning strategies thanks to its intelligent trading algorithm, "LOXM." LOXM was developed by David Fellah, part of J.P. Morgan’s European equity quant research team. Together with the promotion of Chandarana, it can be inferred that usable machine learning tools are better developed by traders and trading desk quants than pure data and machine learning specialists. Other banks may want to take note.
Have a confidential story, tip, or comment you’d like to share? Contact: firstname.lastname@example.org
Bear with us if you leave a comment at the bottom of this article: all our comments are moderated by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. Eventually it will – unless it’s offensive or libelous (in which case it won’t.)
Photo credit: Joe Dunckley/Getty