Jonathan Devor, Developer in Jerusalem, Israel
Jonathan is available for hire
Hire Jonathan

Jonathan Devor

Verified Expert  in Engineering

Data Scientist and Developer

Jerusalem, Israel

Toptal member since May 9, 2022

Expertise
CBig Data ArchitecturePythonWindows DevelopmentLinuxSparkMachine Learning
Bio

Jonathan is an algorithm engineer and data scientist with a firm respect for business needs. He has 10+ years of experience leading projects, from a business need to a working prototype and integration. Jonathan has worked on a wide range of challenges, such as identifying driving violations, detecting online fraud, and optimizing investment strategies. In all cases, he introduced novel approaches and facilitated key breakthroughs. Several of these innovations are now registered as patents.

Portfolio

Mobileye
Python, Spark
RoadMetric (Aquired by Verra Mobility)
Python, Data Integration, Patents
Forter
Big Data, Data Modeling, Bayesian Statistics

Experience

  • Simulations - 10 years
  • Data Science - 10 years
  • Physics - 8 years
  • C - 6 years
  • Numerical Modeling - 5 years
  • Program Design - 4 years
  • Numerical Methods - 3 years
  • Python - 3 years

Availability

Part-time

Preferred Environment

Windows, Linux

The most amazing...

...and important project I've led is an application for automatic detection of drivers who are talking on their phones.

Work Experience

Senior Developer

2021 - 2022
Mobileye
  • Designed and implemented large-scale, cloud-based mapping, which is crucial for the reliable operation of automated vehicles.
  • Developed novel solutions for map labeling given noisy data.
  • Collaborated closely with people within and outside the team to ensure an on-schedule product integration.
Technologies: Python, Spark

Director of Research

2016 - 2021
RoadMetric (Aquired by Verra Mobility)
  • Managed a team of researchers focusing on AI, data science, and advanced image processing.
  • Initiated new products, patents, and standards. RoadMetric was acquired in 2021.
  • Developed solutions to automatically identify drivers who are talking on the phone, measure the speed of surrounding cars via video feed, and track vehicles at night and with limited visibility.
Technologies: Python, Data Integration, Patents

Data Scientist

2016 - 2016
Forter
  • Developed an application to assess the predictive value of specific attributes for identifying fraudulent transactions.
  • Improved the efficiency of model training through pipeline streamlining.
  • Collaborated with analysts to quantify and optimize decision models using historical data.
Technologies: Big Data, Data Modeling, Bayesian Statistics

Senior Quantitative Researcher

2014 - 2015
WorldQuant
  • Constructed novel investment strategies for US and international markets.
  • Implemented a genetic algorithm to automatically create and test machine trading strategies.
  • Made improvements in the foreign exchange simulator.
Technologies: Financials, Trading, Stock Market Techinical Analysis

Experience

Detecting Vehicle Drivers Who Use Mobile Phones

A YOLO-based neural network that identifies video clips of vehicle drivers who are using mobile phones. The application is used to assist police in identifying violations by having them watch only the most-likely video segments.

Measuring Traffic Speeds via Optical Flow

An application that accurately measures 3D velocities of vehicles in traffic by tracing their movements on video streams. Police used this to identify speeding violations using existing stationary traffic cameras.

Increasing Microchip Production Yield by Optimizing Parametric Testing

A novel method that combines results from multiple parametric tests to optimize the criteria for detecting microchip faults. This method is meant to increase microchip production yields in existing pipelines.

A Non-intrusive Method to Measure Direct Electrical Current

While clamp meters can measure electrical currents non-intrusively, they can do so only for alternating currents (AC). I developed and patented a new type of measuring device that can also non-intrusively detect direct currents (DC). It operates by vibrating the sensor and modeling the induced magnetic field near the wire.

Education

2002 - 2008

Ph.D. in Astronomy

Harvard University - Cambridge, MA, USA

1999 - 2002

Bachelor's Degree in Physics

Hebrew University of Jerusalem - Jerusalem, Israel

1999 - 2002

Bachelor's Degree in Computer Science

Hebrew University of Jerusalem - Jerusalem, Israel

Skills

Languages

C, Python, Python 3

Frameworks

Spark

Platforms

Windows, Linux

Storage

Data Integration

Other

Simulations, Physics, Numerical Modeling, Data Modeling, Data Science, Numerical Methods, Program Design, Patents, Big Data, Bayesian Statistics, Stock Market Techinical Analysis, Financials, Trading, Machine Learning, Statistical Methods, Engineering

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