Staff Machine Learning Scientist
Remote, USA
Full-time
About Evoke
AI Agents are rapidly becoming the new operating system for enterprises, yet they are creating a massive, hidden security debt that is compounding daily. At Evoke, we aren’t just watching this shift happen. We are securing it. Our mission is to surface and control the risks that agents pose to businesses. Just as EDR did for endpoints, we are doing for agents with the industry’s first true Agent Detection and Response platform.
About The Role
As a Staff Machine Learning Scientist, you will help us define what separates a safe agent from a dangerous one, then build the solutions that tell the difference. You will create models that recognize intent, drift, and threats as they begin to emerge, enabling Evoke to stop dangerous agents before they can cause any damage.
Primary Responsibilities
Own end-to-end model development, from problem framing and feature engineering through training, evaluation, and production deployment
Help define what best-in-class looks like for agentic threat detection, and set the standards this team measures itself against
Work with other engineers to improve the data that your models learn from, including labeling and ground truth
Handle the realities of security data: severe class imbalance, rare and evolving attacks, ambiguous labels, and adversarial inputs written specifically to evade detection
Architect and build the evaluation suites and production monitoring that make model quality visible and catch drift before a customer reports it
Decide when a model should be retrained, rolled back, or replaced
Work hands-on with engineering to integrate models into production systems and data pipelines
Stay at the forefront of LLMs, applied ML, and AI security techniques, and bring new ideas into the product
Technical Skills
Must Have
7+ years of experience in Data Science/Machine Learning roles
Expert-level applied machine learning, from problem framing and feature engineering through production deployment
Expert-level experience with language models and LLM-based systems, including prompting, evaluation, and fine-tuning
Expert-level training data strategy, including labeling, ground truth, class imbalance, and adversarial examples
Working knowledge of Python and the applied ML stack, large-scale data analysis, and production data pipelines
Nice-to-Haves
Experience in AI security, application security, or threat detection
Backend, developer, or MLOps experience, including model versioning, staged rollout, and rollback in production
Startup experience
Non-Technical Skills
Pragmatic Problem Solving: A focus on practical, deployable solutions that move the product forward. You prioritize resolving customer and platform issues efficiently, while balancing immediate needs with long-term maintainability and support.
Curious & Resilient Mindset: Operates effectively in environments with evolving requirements and incomplete information. Adapts to change, learns new systems and domains quickly, and maintains attention to detail while iterating toward solutions.
Effective Communication: Communicates technical concepts, progress, and blockers clearly to engineers and product partners. Writes well-structured documentation. Helps ensure shared understanding and alignment without unnecessary complexity.
Collaborative Spirit: A commitment to teamwork, empathy, fostering an environment of continuous learning, and shared success. Supports the growth of other engineers by providing thoughtful feedback and helping reinforce good engineering practices.
Results-Oriented: Focuses on delivering high-quality outcomes. Takes responsibility for completing work reliably and follows through on issues and improvements within their scope of ownership.
Compensation
The base salary range for this position in the U.S. is $180,000 - $220,000 per year + equity + benefits.