Machine Learning Engineer

The Opportunity Abnormal Security’s Account Takeover team is pioneering the future of security for Software as a Service and cloud offerings. Corporations of all sizes have started to utilize cloud services from Google Docs to Box to Slack, shifting work from confined office buildings or firewalls to the open web. Cybercriminals are very aware of the opportunities to steal data, hijack financial information, and compromise companies that use these cloud services. We need your help to build a fresh layer of protection that will offer enterprises the same level of security for their cloud offerings as Abnormal Security’s leading products in the email security domain. We’re seeking to recruit a Machine Learning Engineer to join our Account Takeover (ATO) Detection team. The objective of the ATO Detection team is to deliver top-notch attack detection efficacy to combat the rapidly evolving attack landscape using a mix of generalisable and auto-trained models, along with specific detectors for high-value attack categories. This team is addressing a complex detection issue - from modeling communication patterns to establishing enterprise-wide baselines, to normalising across multiple event sources, and using contextual information to avoid false positives. This role strikes a balance between velocity and excellence. About You You’re someone who wants to make a difference. You’re committed to solving customer problems and have a budding set of skills in machine learning, software engineering, and data science. You’re eager to apply these skills to problems that will make the world a better place. You’re humble and open to learning! As one of your initial jobs - possibly your very first job - you know that there’s a vast array of skills to develop and knowledge to gain, and you’re determined to do so as swiftly as possible. You ask questions. You make notes. You take an active and curious approach to your work and consequently grow faster than the average individual. We're a determined team: we're building a new product from the ground up - this implies that you should be comfortable with a level of uncertainty beyond what you'd find at a more established company or even a more established team. Not every project will come with a well-defined project requirement document - that's expected and we foresee this engineer diving in to figure out what to do. This position demands a certain skill set Skills/Experience - Required: Excellent software engineering skills, strong computer science fundamentals, fluency with Python and machine learning libraries like numpy and scikit-learn, familiarity with data processing frameworks like Pandas and Spark, a systematic approach to debug both data and system issues with ML models or heuristics, writing code that is easily testable and comprehensible for other engineers. A machine learning academic background (Bachelor's degree in Computer Science or related fields), hands-on experience training and tuning models, 1+ years of experience or 2+ internships to develop these skills in a production environment, and Interest in security and stopping bad actors. Skills - Nice to have: Experience with tuning a machine learning system in a production setting, Master’s in Computer Science or related field, experience working in a startup environment, familiarity with LLMs. This position is not: A role focused on optimizing existing machine learning models, a research-oriented role that's disconnected from the product or customer, a statistics/data science meets ML role. Role Responsibilities As a Machine Learning Engineer on the ATO team, you will: Design and build systems that blend rules, models, feature engineering, and business and product inputs into an ATO detection product, guided by senior engineers, Build attack detection systems capable of highlighting rare, suspicious activity with 95%+ precision and less than 1 minute latency on the event stream, understand the nature of attacks and design features to calibrate behavior across our diverse customer base, write testable, readable code that handles feature drifts between online/offline data, contribute to other areas of the stack when needed, participate in building a world-class detection engine across all layers, work with infrastructure and systems engineers to develop appropriate feature aggregates for the detection system, and create a memorable work environment for colleagues and interviewing process for candidates. #LI-ML1