Machine Learning Engineer — Wearable Biosignals

Part-Time W-2 Position Initially | Employment Contingent Upon NIH SBIR Funding

Lumily Health | Richmond, Virginia preferred; U.S.-remote candidates considered | Approximately 20 hours per week

Lumily Health is an early-stage pediatric health technology company researching whether wearable sensor data and machine learning can reveal meaningful physiological patterns during children’s recovery at home after hospitalization.

Our proposed research sits at the intersection of wearable sensing, physiological time-series analysis, machine learning, and pediatric care. Lumily is exploring approaches involving individualized physiological baselines and multisignal analysis, including patent-pending methods.

Lumily’s technology is investigational and is not currently available for clinical use. It has not been cleared or approved by the FDA to diagnose, predict, monitor, prevent, or treat any medical condition.

We are preparing an NIH SBIR Phase I application and are selecting a machine-learning engineer to join the proposed technical team if the project is funded.

We care deeply about the problem we are working to solve, but we also believe building something meaningful should be energizing and genuinely enjoyable. We are a collaborative, curious, low-ego team doing technically ambitious work—and we plan to have a lot of fun while we do it.

This is a future, grant-funded position rather than an immediate-start opening.

The selected candidate will be included in Lumily’s NIH SBIR Phase I application as proposed project personnel. Before submission, the candidate will be asked to confirm their good-faith intention to participate if the award is made, sign a conditional letter of commitment, and provide the required NIH application materials.

Employment will begin only if Lumily receives the NIH award and completes final employment and onboarding documentation.

Because this role will perform core research and development as part of Lumily’s internal technical team, it must be structured as a part-time W-2 employee position. A 1099 or independent-contractor arrangement is not available.

This opportunity may be particularly well suited to an experienced machine-learning engineer seeking meaningful, mission-driven part-time work and the opportunity to help shape an early-stage pediatric health technology company.

The ideal candidate can contribute independently at approximately 20 hours per week during the initial Phase I project and is interested in the possibility of expanding into a larger or full-time role as Lumily’s funding, research program, and technical team grow.

You may especially enjoy this role if you like:

• Solving unusual and genuinely difficult technical problems.
• Working closely with a founder and having a real voice in technical decisions.
• Helping shape the earliest stages of a product and research program.
• Exploring new approaches rather than maintaining mature systems.
• Working with smart, kind people who take the mission seriously without taking themselves too seriously.

• Develop and evaluate machine-learning approaches for multimodal physiological and wearable time-series data.
• Explore methods for individualized baseline formation, longitudinal pattern analysis, anomaly identification, and multisignal data fusion.
• Build reproducible research pipelines for sensor-data ingestion, preprocessing, signal-quality assessment, feature development, modeling, and evaluation.
• Investigate challenges common to wearable datasets, including motion artifacts, missing data, sensor noise, interpatient variability, and limited clinical sample sizes.
• Help translate Lumily’s proposed research aims into a rigorous computational and model-evaluation plan.
• Support the assessment of technical feasibility within the scope of the proposed Phase I research.
• Collaborate with Lumily’s clinical and research partners, hardware contributors, and statistical advisors.
• Document methods, assumptions, experiments, limitations, and findings for NIH reporting and future research planning.
• Work directly with Lumily’s founder and principal investigator as part of a small, fast-moving team where your ideas and technical judgment will matter.

• Strong applied machine-learning experience involving time-series, sensor, physiological, or other complex longitudinal data.
• Advanced proficiency in Python and relevant tools such as PyTorch or TensorFlow, scikit-learn, pandas, NumPy, and scientific-computing libraries.
• Experience designing reproducible experiments and selecting appropriate evaluation methods and performance metrics.
• Understanding of small-dataset modeling, overfitting, data leakage, uncertainty, missing data, and class imbalance.
• Experience moving research concepts into functional prototypes or experimental pipelines.
• Ability to communicate technical decisions, findings, and limitations clearly to clinical, research, and product stakeholders.
• Comfort working in an early-stage research environment with incomplete data and evolving technical requirements.
• A collaborative, curious, and practical approach to solving hard problems.
• Authorization to work as a W-2 employee in the United States.
• Ability to perform all grant-supported work while physically located within the United States.

Experience in every area below is not required. We are especially interested in candidates with exposure to one or more of the following:

• Wearable devices, physiological sensing, or health-related longitudinal datasets.
• Heart rate, heart-rate variability, respiration, temperature, motion, audio, electrodermal activity, or other biosignal data.
• Signal processing, anomaly detection, change-point analysis, personalized modeling, or multimodal data fusion.
• Digital health, clinical research, pediatric health, or regulated medical-technology development.
• HIPAA-aligned data environments, IRB-governed research, or clinical datasets.
• NIH, SBIR/STTR, or other federally funded research.
• Publications, open-source work, or experimental projects involving physiological or longitudinal data.

Direct pediatric experience is helpful but not required for candidates with strong physiological time-series or wearable-sensing experience.

• Employment type: Part-time W-2 employee initially.
• Anticipated commitment: Approximately 20 hours per week during the initial NIH SBIR Phase I project period.
• Initial grant-funded period: Approximately six months, subject to the final Notice of Award and approved project period.
• Compensation: $65 per hour.
• Start date: Employment will begin only if Lumily receives the proposed NIH SBIR Phase I award and completes final employment documentation.
• Location: Richmond, Virginia preferred. Remote candidates who will perform all grant-supported work within the United States will be considered.
• Growth potential: The position will begin part-time, with the potential to expand into a larger or full-time role as Lumily’s funding, research program, and technical needs grow.
• Team environment: Collaborative, candid, curious, low-ego, and fun. We are doing serious work, but we believe building it should be exciting too.

Before the NIH application is submitted, the selected candidate will be asked to:

• Confirm their good-faith intent and anticipated availability if the award is made.
• Review and agree to the proposed project responsibilities and estimated level of effort.
• Authorize Lumily to include them in the NIH application.
• Complete the required NIH biosketch materials through SciENcv.
• Provide required application disclosures and documentation.
• Sign a conditional letter of commitment.

These steps do not begin employment and do not require the candidate to leave or modify their current position.

Lumily will not expect the candidate to perform unpaid engineering, model development, data analysis, or other substantive project work before employment begins.



Please submit:

• A resume or CV.
• A brief note describing your relevant machine-learning, time-series, biosignal, or wearable-data experience.
• Links to one or two relevant projects, publications, repositories, or technical examples, when available.
• Confirmation that you anticipate being available for approximately 20 hours per week if the project is funded.
• Confirmation that you can perform the work while physically located within the United States.

Formal cover letters are not required. We would rather hear—in your own words—what interests you about the work, what you have built, and why you think you might enjoy building with us.

Lumily Health provides equal employment opportunities to qualified applicants and employees in accordance with applicable law.