Why COMO
COMO is building the future of cyber resilience. Our AI-native platform helps organizations understand their ability to recover from cyberattacks, identify critical gaps and dependencies, and prepare for what happens after an attack. Backed by a significant seed round led by General Catalyst, with a live product and paying customers, we're building an exceptional team to define a new category in cybersecurity.
About the Role
We're looking for a Data Scientist to solve one of the hardest problems in our product: grouping thousands of cloud assets into the applications they actually belong to, and ranking which ones matter most, from graph structure, tags, naming, and usage signals, with no labeled data to start from. This is what lets COMO test recovery at the level customers care about: whole applications, not single assets. Beyond the models, you'll be the person who measures whether the product is actually getting better, and makes sure we do it at a cost that scales. Your work will be measured on real environments of some of the largest enterprises in the world.
What You'll Do
Build the application-classification engine: combine graph structure, tags, naming, and network signals into application groupings with confidence scores
Build asset-criticality classification: rank what matters by production-ness, blast radius, and usage
Own measurement: define how we know our models and our product are good, with the statistics to back it, and report it as a number that drives decisions
Design the feedback loop: turn customer corrections and manual overrides into training signal
Keep an eye on cost: design models and pipelines that stay efficient as data volume grows across customers
Ship models into the product via API or pipeline, and work closely with the engineers behind our dependency graph and AI platform
What We're Looking For
M.Sc. or higher in Data Science, Statistics, Computer Science, Mathematics, or a related quantitative field
4+ years of experience in a Data Science or ML role, with at least one production system doing entity resolution, clustering, or classification over messy real-world data
Strong research and statistics foundations: experiment design, evaluation methodology, and knowing when a result is real
Experience defining and owning quality metrics for a model or product, and using them to drive decisions
Experience shipping models into a product (API or pipeline), not only analyses
Strong feature engineering over graph and relational signals, comfortable with both embeddings and classic ML
Strong SQL and Python data-stack skills, with cloud-native experience
Understanding of cost optimization for data and ML workloads at scale - advantage
Experience with infrastructure or operational data (asset inventory, service maps, security graphs) - advantage
Security, observability, or IT-asset domain background - advantage
Our Stack
Python, Polars, PostgreSQL, Kafka, AWS Bedrock, with an AI-native workflow throughout.