Emmanuel Tsukerman, Developer in Jerusalem, AR, United States
Emmanuel is available for hire
Hire Emmanuel

Emmanuel Tsukerman

Verified Expert  in Engineering

Machine Learning Developer

Location
Jerusalem, AR, United States
Toptal Member Since
January 13, 2020

Emmanuel graduated from Stanford University. In 2017, his ML-based anti-ransomware product was awarded Top 10 Ransomware Products of 2018 by PC Magazine. In 2018, he designed an ML-based malware detection system for Palo Alto Network's WildFire service (over 30,000 customers). In 2019, Emmanuel authored the Machine Learning for Cybersecurity Cookbook and launched the Cybersecurity Data Science Course on Udemy.

Availability

Part-time

Preferred Environment

Git, Jupyter, PyCharm, MacOS

The most amazing...

...project I've developed is an anti-ransomware product built from scratch that utilized machine learning to detect Ransomware viruses before they execute.

Work Experience

Data Scientist

2018 - 2019
Palo Alto Networks
  • Completed R&D of next-generation machine learning anti-malware system for WildFire production system (>30,000 enterprise users).
  • Initiated, researched, and developed a machine learning system for malicious PowerShell script detection.
  • Contributed to R&D for an AI-based NLP interface for the Panorama network security management system.
Technologies: Hadoop, Keras, Scikit-learn, Python

Data Scientist

2017 - 2018
CyberSight Inc.
  • Completed R&D of machine learning-based anti-ransomware solution in a small startup.
  • Won Top 10 Best Ransomware Protection of 2018, Editor’s Choice (PC Magazine).
  • Developed machine learning features for an anti-ransomware solution.
  • Performed business analytics using Splunk.
Technologies: Splunk, Boost, C++, Scikit-learn, Python

Machine Learning Engineer

2015 - 2016
SimInsights Inc.
  • Implemented online partially observable Markov decision process (POMDP) solver for personalized learning.
  • Implemented inverse reinforcement learning solution for personalized learning.
  • Created mathematical modeling for infectious disease propagation simulation for Tufts University.
Technologies: Python

Team Lead

2011 - 2011
UCLA, Research in Industrial Projects for Students
  • Led team of four in implementing kernel density estimation (KDE) algorithm for crime prediction. The product was utilized by LAPD in the Topanga district for motor vehicle burglary prevention.
Technologies: MATLAB

CyberSight RansomStopper

https://www.pcmag.com/review/357847/cybersight-ransomstopper
I developed the product RansomStopper from conception to market, including its machine learning capabilities. There are two versions, a consumer and an enterprise version. Here's what PC Magazine had to say about the consumer version:

"CyberSight RansomStopper offers free, dedicated ransomware protection, and it now handles ransomware that launches at Windows startup. It's a winner, and free."
- Neil J. Rubenking (PC Magazine product reviewer) June 21, 2018

Editors' Rating:4.5/5, excellent

Machine Learning for Cybersecurity Cookbook

A handbook for the practitioner of machine learning in the field of cybersecurity, authored with publisher Packt.

The most up-to-date most cutting-edge most practical guide out there for applying ML/data science to security!

It teaches:

1. Smart malware detection and evasion techniques using ML, deep learning for catching zero-day threats, overcoming obfuscation/packing, automating malware analysis...

2. Social engineering using ML including voice impersonation, Deepfake, and fake review generation and *phishing on steroids* via ML...

3. Next-gen pentesting including NN-based fuzzing, Metasploit made DANGEROUS with an RL agent, software vulnerability detection with AI and Tor deanonymizing!

4. Upgraded intrusion detection with AI including Insider Threat detection, network anomaly detection, catching DDoS and financial fraud.

5. Securing and attacking data with ML including Deep Learning for password cracking (#scary), steganalysis via AI, ML attacks on hardware, and encryption with NNs.

6. Secure and Private AI to preserve customer privacy and confidentiality, including federated learning, differential privacy, and even adversarial robustness.

Machine Learning for Red Team Learning Path, InfoSec Institute

https://www.infosecinstitute.com/skills/learning-paths/machine-learning-for-red-team-hackers/
I've developed the Machine Learning for Red Team learning path in collaboration with the InfoSec Institute.

Here's the copy:

Everyone knows that AI and machine learning are the future of penetration testing. Large cybersecurity enterprises talk about hackers automating and smartening their tools; The newspapers report on cybercriminals utilizing voice transfer technology to impersonate CEOs; The media warns us about the implications of DeepFakes in politics and beyond...

This course finally teaches you how to do and defend against all these things.

This course will be teaching you, in a hands-on and practical manner, how to use machine learning to perform penetration testing attacks, and how to perform penetration testing attacks on machine learning systems.

You will learn
• How to supercharge your vulnerability fuzzing using machine learning.
• How to evade machine learning malware classifiers.
• How to perform adversarial attacks on commercially-available machine learning as a service model.
• How to bypass CAPTCHAs using machine learning.
• How to create deep fakes.
• How to poison, backdoor and steal machine learning models.

And you will solidify your slick new skills in fun hands-on assignments.

I wish this course was for everyone but unfortunately, it just ain’t so. You should enroll only if you are really passionate about computer security and want to be the best at what you do. This course will challenge you and introduce you to new ideas. It will offer you fun hands-on assignments that will require you to bypass CAPTCHA challenges, get your hands dirty and modify malware, fuzz a secretly-vulnerable program, trick a commercially-available machine learning as a service and create a realistic fake video. If this sounds exciting for you, then click the enroll button to get started!

Cybersecurity Data Science Learning Path, InfoSec Institute

https://www.infosecinstitute.com/skills/learning-paths/cybersecurity-data-science/
Authored online cybersecurity data science training available at InfoSec Institute.

Here's the copy:

This hands-on, comprehensive skill path covers everything from the fundamentals of cybersecurity data science to the state of the art. Among many other practical lessons, you will be setting up a cybersecurity lab, constructing classifiers to detect malware, utilizing deep learning technology and even hacking security systems with the help of machine learning — all taught by an award-winning expert in the field of cybersecurity data science.

Languages

Python, Python 3, SQL, C++, R, JavaScript

Libraries/APIs

Keras, Scikit-learn, TensorFlow, PyTorch

Paradigms

Data Science, Agile Software Development

Other

Machine Learning, Malware Analysis, Software Development, Artificial Intelligence (AI), IT Security, Natural Language Processing (NLP), GPT, Generative Pre-trained Transformers (GPT)

Frameworks

RStudio Shiny, Boost, Hadoop, Spark

Tools

Splunk, PyCharm, Jupyter, Git, MATLAB, Microsoft Power BI, Tableau

Platforms

MacOS, Amazon Web Services (AWS), Docker

2013 - 2017

Ph.D. in Applied Mathematics

UC Berkeley - Berkeley, CA

2012 - 2013

Master's Degree in Computational and Mathematical Engineering (CME)

Stanford University - Palo Alto, CA

2009 - 2013

Bachelor's Degree with Honors in Mathematics

Stanford University - Palo Alto, CA

MARCH 2018 - PRESENT

Deep Learning Specialization

Coursera

DECEMBER 2016 - PRESENT

IT Fundamentals

CompTIA

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring