Summary
Join Abnormal Security as an Applied Data Scientist to contribute to building a high-recall detection engine for message attacks, utilizing machine learning and behavioral AI.
Requirements
- 5+ years experience designing, building product machine learning applications in one of the domains of text understanding, entity recognition, NLP experience, computer vision, recommendation systems, or search
- Experience with data analytics and wielding SQL+pandas framework to both build metric and evaluation pipelines, and answer critical questions about counterfactual treatments
- Ability to understand business requirements thoroughly and bias toward designing a simplest yet generalizable ML model / system that can accomplish the goal
- Ability to rapidly iterate on 0-to-1 model prototypes, interpret results, and pivot an approach, in order to evaluate most promising solutions as new problems arise
- Uses a systematic approach to debug data issues within both ML and heuristics models
- Fluent with Python and machine learning toolkits like numpy, sklearn, pytorch and tensorflow
- Effective programming skills which enable them to quickly add incremental logic to our codebase with readable, well tested and efficient code
- BS degree in Computer Science, Applied Sciences, Information Systems or other related engineering field
Responsibilities
- Deep inspection and row level data analysis of our false negatives and false positives
- Produce data and feature insights to iteratively improve our detection efficacy
- Understand features that distinguish safe emails from email attacks, and utilize them effectively into our models stack and engine
- Train models and develop detectors on well-defined datasets to improve model efficacy on specialized attacks
- Identify and recommend new features groups or ML model approaches that can significantly improve detection efficacy for a product
- Work with infrastructure & systems engineers to productionize signals to feed into the detection system
- Writes code with testability, readability, edge cases, and errors in mind
- Actively monitor and improve FN rates and efficacy rates for our message detection product attack categories
- Contribute in other areas of the stack: building and debugging data pipelines, or presenting results back to customers in our tools when the occasion arises