Talks #
Every now and then, I find myself in front of an audience.
2024 #
Practical Considerations for Machine Learning in Fraud Prevention Programs #
April 25, 2024
Boston, MA
Event: Open Data Science Conference (ODSC) East
In this presentation, we will walk through in general terms how machine learning has been utilized in a financial technology company in the web3 space to detect and prevent fraud. We will cover the particular considerations of fraud as it relates to building and deploying effective machine learning models. We will lightly delve into the code and tooling to illustrate from a practical standpoint how the machine learning system can be constructed. At the end of the presentation, the audience should have a conceptual understanding of how a machine learning program can be implemented to prevent fraud.
The effective application of machine learning to detecting fraud does require significant nuance and measured consideration as there are unique attributes to fraud that are not present in other domains, including: the delayed availability of credible labels for fraud (sometimes fraud labels are not available for months after particular events, if ever), variations of the types of fraud that might manifest (Is it first party fraud or third party fraud? Is it truly fraud, or is it merely unfavorable customer behavior?), and considerations around timeliness to be able to prevent fraud rather than to merely react to fraud after the fact, at which point recourse options for a business might be limited.
In addition to the domain concerns of how machine learning fits into a broader fraud prevention program, there are practical considerations extending from how fraud detection models are trained, to the machine learning ops considerations that are necessary to serve and maintain timely and reliable models, and through to the support mechanisms necessary to ensure that machine learning is persistently available as a business-critical function.
2022 #
A Macroscopic Review of Cloud Exploits and Exposures #
March 9, 2022
10:40 AM – 11:30 AM
Boston, MA
Event: SecureWorld Boston 2022
A systematic review of some of the most prominent cloud exploits in 2021, as well as an assessment of internet-wide telemetry collected across the entirety of the IPv4 space to identify exposures that could hint at opportunistic targets within cloud infrastructures.
In this talk, we’ll take a moment to systematically review some of the most prominent cloud exploits in 2021 that have since been publicly disclosed. We’ll cover details about the industry distributions that were known to be harmed, the types and scale of exposures that occurred, and the underlying factors that contributed to exposure.
We’ll also supplement that review of the state of cloud security in 2021 with an assessment of internet-wide telemetry collected across the entirety of the IPv4 space to identify exposures that could hint at opportunistic targets within cloud infrastructures.
With this knowledge, organizations can take focused, proactive measures to mitigate the risks facing cloud implementations.
2019 #
Identifying Cyber Exposure Through Internet Telemetry #
October 25, 2019
Raleigh, NC
A presentation on using data visualization methods to understand an overwhelming volume of Internet telemetry.
Reflecting on Cyber Risk with Internet Telemetry #
September 27, 2019
Natixis Compliance and Risk 2019
Presentation on recent cybersecurity concerns premised on broadly-collected Internet telemetry for an audience of lawyers and investment managers.
2018 #
An Introduction to Applying Data Science to Security #
May 19, 2018
3:30 PM – 4:15 PM
A talk I co-presented with Vasudha Shivamoggi on about using data science methods to analyze Internet telemetry collected throught IPv4 scanners and honeypots.
Data science as a multi-disciplinary field is in a fairly nascent state, but it is burgeoning and beginning to find its way into a wide array of industries, often proving to be highly impactful. Data science approaches may sound unfamiliar to security practitioners, but can be tremendously effective in addressing challenges in the security space, particularly in detecting and responding to security incidents.
In this session, we will introduce the audience to a series of common steps that are essential to implementing a data science approach: acquiring security and network data, manipulating it into a suitable structure for analysis, identifying and applying a range of appropriate analytical methods based on the scenario, assessing the utility and reliability of the results, and communicating the findings to stakeholders in a digestible manner.
The audience will gain exposure to applying a systematic lens to analyzing security-related data, which differs markedly from the more traditional case-based approach to addressing security challenges.
Precisely Predicting Churn: Find Churn Faster with Machine Learning #
March 14, 2018
2:00 PM – 3:00 PM
Salt Lake City, UT
A talk I co-presented on about combining data science methods with business intelligence dashboarding to predict customer churn.
A talk on using using machine learning methods to predict customer churn, deploying the model through AWS, and presenting the findings internally using the Domo business intelligence platform.
Actionable UX Analytics with Pendo #
January 18, 2018
6:00 PM – 8:00 PM
Pendorama Boston 2018
A workshop presentation on analyzing customer activities on-site using machine learning methods to determine how to improve the overall customer experience.