We're growing our Data Science team to ship production machine learning that powers ActivTrak's next generation of workforce intelligence products. This is a hands-on role built around production ownership: you take a problem from formulation through a shipped, monitored model, not just to a notebook.
You'll work on problems like:
- Predicting user roles and classifying activity from behavioral event data
- Cross-account benchmarking that turns aggregate usage patterns into product-differentiating insight
- Making pragmatic tradeoffs between model sophistication, business value, reliability, latency/cost, and iteration speed
- Partnering with our Data Engineers, who take your models and analysis and turn them into durable, scalable production systems
Where this role starts and ends: You own problem formulation, feature definition, model and scoring logic, evaluation, and ongoing model performance in production. Data Engineering owns the pipeline infrastructure, orchestration, deployment mechanisms, and operational reliability that put your work into production. The primary ownership is clear, but you'll work together across that boundary when production issues span model and platform — your job is the model delivering the intended outcome and improving over time.
This is an individual-contributor role with substantial ownership over your problem space. It does not include people-management responsibilities.
Requirements
Must-Haves:
- 5+ years bringing machine learning to production at scale, with direct experience making the sophistication/speed/reliability tradeoffs described above
- Strong Python and production-grade software engineering practices (testing, code review, version control)
- Experience with classification/prediction problems on behavioral, event, or user activity data
- Strong SQL and comfort working directly against production data sources, not just flat files or CSVs
- Experience with feature engineering: defining, standardizing, and validating features for production models
- Experience monitoring model quality after deployment and responding to drift or degradation, not just shipping and moving on
Nice-to-Haves:
- Time series analysis and/or hidden state models
- Parallel dataframes (Dask, Spark, or similar)
- Comfort working within a layered/medallion-style data architecture (raw → cleansed/identified → aggregated/de-identified)
- Feature store experience (versioning, storage, reuse)
- Cloud environment experience (GCP or AWS), and general comfort operating around containerized/orchestrated infrastructure (Docker, Kubernetes) even if you're not the one building it
Benefits
Why Should You Apply?
- Own production ML that reaches customers and improves through real-world feedback
- Work on a genuinely uncommon ML problem: behavioral event data from 9,500+ customer organizations, used for role prediction, activity classification, and cross-account benchmarking
- Small, senior team with real ownership and visibility to leadership
Work environment
- Position is remote within US
- Minimal travel
- Limited physical demands
This is an incredible opportunity to embark on an exciting journey with a dynamic, VC-backed company. If you have a proven track record of creative thinking, a drive for learning, and a deep commitment to collaboration, we want to talk to you!
ActivTrak is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. ActivTrak does not discriminate on the basis of race, color, religion, sex, national origin, political affiliation, sexual orientation, marital status, disability, age, protected veteran status, gender identity or any other factor protected by applicable federal, state or local laws.
