Matthew Darnall

Current

• Associate at Credit Suisse

Past

• Associate at Credit Suisse
• Laborer at Kernen Construction
• Intern at Motorola

see less...

Education

• University of Wisconsin-Madison
• Humboldt State University

Connections

46 connections

Industry

Financial Services

Matthew Darnall’s Summary

Work in a job with that is interesting (i.e. uses non-trivial math and serves some purpose).

Matthew Darnall’s Specialties:

math finance, F#, a little C++

Matthew Darnall’s Experience

• Associate

  Credit Suisse

  (Public Company; CS; Banking industry)

  July 2008 — Present (1 year 5 months)

  I work on Interest Rate Products in the Global Modelling and Analytics Group. I mostly write code to value some exotic products, develop trading tools, and do some general quant / coding work as it is needed.

• Associate

  Credit Suisse

  (Public Company; CS; Banking industry)

  May 2007 — August 2007 (4 months)

  Summer intern, I did a project on a paper of Carr's about statically hedging with Credit Default Swaps

• Laborer

  Kernen Construction

  (Construction industry)

  May 2004 — July 2006 (2 years 3 months)

  Worked here during the summers and a couple breaks during undergrad. Mostly I did basic industrial/commercial carpentry, poured concrete, dug some sewer lines, ran piping, a little paving, etc. I was mostly there as cheap, unskilled labor so I did the shoveling, carrying stuff, tying rebar, etc.

• Intern

  Motorola

  (Public Company; MOT; Telecommunications industry)

  May 2005 — August 2005 (4 months)

  I worked on cryptography here, particularly implementation of AES on an ARM7TDMI processor, which Motorola used at the time. I presented this work (with Doug Kuhlman at Motorola) at the IndoCrypt conference.

Matthew Darnall’s Education

• University of Wisconsin-Madison

  PhD , Math , 2004 — 2008

  Wrote a thesis on low discrepancy sequences. Before me, nobody knew that a random (0,m,2) net (with 2^m points, of course) would have a L^\infty discrepancy function that is normally distributed with variance on the order of m (i.e. the log of the number of points). I gave an extension of this result to higher dimensions, and proved that the discrepancy of (t,m,s) nets in dimension higher than 2 are NOT normally distributed (they have fat tails). Too bad, if they were normally distributed, the "Great Open Conjecture" in low discrepancy sequences would be false. BTW, I still think it is false, but that is just me and a couple of harmonic analysts I have talked to :).

  Activities and Societies:

  Face Recognition Group, Number Theory Group, Cryptography Group

• Humboldt State University

  BA , Mathematics , 2002 — 2004

  Activities and Societies:

  CASA

Additional Information

Matthew Darnall’s Groups:

•    Wisconsin Alumni Association
•    Humboldt State University Alumni Association

Matthew Darnall’s Contact Settings

Interested In:

• reference requests
• getting back in touch