Skip to content
  • Overview
  • Team
  • Capabilities
  • Projects
  • Blog
  • Demos
  • Presentations
  • Publications
  • Sophos.com

Joshua Saxe

Chief Scientist

Joshua Saxe is Chief Scientist at Sophos, where he manages the Sophos AI team, and provides strategic leadership for the company’s research and development efforts.  Before joining Sophos, Joshua was Chief Data Scientist at Invincea, where he led the development of machine learning systems for analyzing, detecting and blocking malware executables and malicious behavior, and was Principal Investigator on multiple DARPA funded efforts focused on applying machine learning and data visualization to the federal government’s cybersecurity challenges.  He is the author, with Hillary Sanders, of the book Malware Data Science, from No Starch Press, and an author on multiple patents and peer-reviewed security machine learning papers.

@joshua_saxe
Joshua Saxe' Profile Image'

Blog Posts

View all of Joshua’s blog posts
October 5, 2020October 5, 2020

Lessons learned from building a 4,000+ member cybersecurity volunteer organization in four months

When I posted this tweet in March of this year, kicking off a process which would give birth to the […]

Joshua Saxe
May 8, 2020June 19, 2020

Detecting Malicious URLs and Stopping the Attack Early

Any good attack-chain usually involves tricking users at some point, whether it’s getting them to run a malicious file because […]

Alex Long
Joshua Saxe
Konstantin Berlin

Presentations

View all of Joshua’s presentations
August 9, 2020August 12, 2020

DEF CON 28 AI Village: Detecting hand-crafted social engineering emails with a bleeding-edge neural language model

Joshua Saxe
Younghoo Lee
July 23, 2017February 19, 2021

The New Cat and Mouse Game: Attacking and Defending Machine Learning Based Software

Joshua Saxe

Publications

View all of Joshua’s publications
August 24, 2021

CatBERT: Context-Aware Tiny BERT for Detecting Targeted Social Engineering Emails

Targeted phishing emails are a major cyber threat on the Internet today and are insufficiently addressed by current defenses. In this paper, we leverage industrial-scale datasets from Sophos cloud email security service, which defends tens of millions of customer mailboxes, to propose a novel Transformer-based architecture for detecting targeted phishing emails. Our model leverages both natural language and email header inputs, is more computationally efficient than competing transformer approaches, and we show that it is less prone to adversarial attacks which deliberately replace keywords with typos or synonyms.

Younghoo Lee
Joshua Saxe
Richard Harang
May 22, 2020

Garbage in, garbage out: how purportedly great ML models can be screwed up by bad data

Hillary Sanders
Joshua Saxe
May 22, 2020May 22, 2020

SeqDroid: Obfuscated Android Malware Detection Using Stacked Convolutional and Recurrent Neural Networks

Richard Harang
Younghoo Lee
Joshua Saxe

Patents

  • Methods and apparatus for detecting malware samples with similar image sets Patent # 9,672,358
Sophos AI - Smarter Security
Follow Us
  • Facebook
  • Instagram
  • LinkedIn
  • RSS
  • Twitter
  • YouTube
  • Legal
  • Privacy
  • Cookies
© 1997–2022 Sophos Ltd. All rights reserved.