Machine Learning Engineer— Wearable Sensor Data

About Striv

Striv is building wearable performance technology that turns high-frequency movement data into actionable insights for athletes. Our system captures rich raw sensor streams from the body in motion, then transforms them into reliable metrics, models, and product experiences. We are a small, high-caliber technical team with backgrounds across places like MIT, Harvard, Columbia, and leading engineering/product organizations. We move quickly, but we are disciplined about scientific validity, product quality, and honest technical judgment. The team is flexible, collaborative, and low-ego — we care about doing strong work with people who are thoughtful, kind, and easy to work with.

The Role

We are looking for a hands-on data and machine learning engineer to work closely with our raw sensor data and help build the analytical foundation behind Striv’s product.

You will dig into high-frequency time-series signals, develop analysis pipelines, validate metrics, debug model behavior, and train machine learning models that translate messy real-world movement data into meaningful product intelligence. Over time, this role can grow into our broader effort to build representation learning and foundation models for human movement from raw sensor data.

We care more about judgment, rigor, and ability to learn than years of experience. Strong early-career candidates are welcome, especially if you have real experience working with messy data and can think clearly from first principles.

We are especially interested in people who sit at the intersection of data, machine learning, and consumer wearable, biometric, or health data systems. Experience with products or datasets similar to WHOOP, Oura, Garmin, Apple Watch, Fitbit, or other real-world consumer health and human data systems would be highly relevant.

What You’ll Do

Identify signal patterns, artifacts, failure modes, and sources of noise in real-world movement data.

Build analysis pipelines that turn raw data into clean, reliable features and datasets for modeling.

Train, evaluate, and debug machine learning models on time-series sensor data.

Support our long-term work in self-supervised learning, representation learning, and modeling.

Validate metrics carefully before they become part of the product experience.

Question results honestly and investigate why a model, metric, or number behaves the way it does.

What We’re Looking For

Master’s degree or above in Computer Science, Machine Learning, Data Science, Statistics, Electrical Engineering, Biomedical Engineering, Biomechanics, or a related technical field.

1–3\+ years of relevant hands-on experience with data, machine learning, signal processing, or time-series analysis.

Experience with wearable sensors, biometric data, physiological signals, health data, sleep/recovery data, or real-world human data systems is a strong plus.

Strong Bachelor’s candidates may be considered only if they have 3–5\+ years of highly relevant experience or substantial hands-on work with real-world sensor, time-series, or ML systems.

Strong hands-on ability with data: you like opening raw files, plotting signals, and understanding what is really happening.

Solid fundamentals in machine learning, statistics, and experimental thinking.

Good instincts for working with small, noisy, biased, or imperfect datasets.

Strong Python skills and comfort with NumPy, pandas, plotting tools, and modern ML workflows.

Experience with at least one deep learning framework, ideally PyTorch.

Ability to take an ambiguous technical problem, structure it, and make progress independently.

Comfortable using modern AI coding tools such as Claude Code, Codex, or similar systems.

Benefits

Competitive salary and equity.

Health insurance.

401(k) plan.

Flexible remote-friendly work environment.

Opportunity to work directly on real-world human movement data, wearable sensors, and applied ML systems.

Small, high-caliber team with high ownership, fast product iteration, flexible working style, and a supportive, low-ego culture.