Head of Science (Algorithms)

About the role:

Luna builds wearables that turn raw body signals into things people can act on. We are building

LifeOS, the layer that connects what someone does with how their body responds, and

separates real cause and effect from coincidence. We are hiring a senior scientist to own that

layer and to set the scientific and algorithmic standard across every product we ship.

This is a hands-on leadership role. You will build, validate and ship, and lead a small team of

scientists and algorithm engineers as it grows.

What you will do:

Own the science of LifeOS: the causal and correlational engine that turns wearable,

behavioural and context data into recommendations that measurably move a health

outcome.

Design and run causal inference methods for noisy, observational, individual-level (N-of-1)

data, and know where the data cannot support a claim.

Lead algorithm development across the company: heart rate, HRV, sleep, activity, recovery

and derived scores from PPG and motion signals.

Set one validation standard: reference-based accuracy testing (ECG, PSG, lab), benchmark

datasets, release gates and regression tracking, so accuracy is measured and tracked, not

argued.

Design and run validation studies and clinical protocols, including ethics approvals, and

publish where it builds credibility.

Work with firmware, app, data and product teams so algorithms move from research to

production reliably.

Define how we test whether an actionable works: experiments, interventions and outcome

measures.

Track regulatory boundaries (wellness versus medical claims) and shape what we can say

about our features.

Hire, mentor and raise the bar for the science and algorithms team.

What we are looking for:

PhD in biomedical signal processing, computational health, causal machine learning,

biostatistics, or a closely related field, from a strong research university in India or abroad.

An MS with deep industry track record will be considered.

5 to 8 years of experience after the PhD, in industry or in an applied lab with real product

output.

Demonstrated depth in causal inference on observational data: DAGs, instrumental

variables, g-methods, Bayesian and time-series causal models, or equivalent.

Shipped algorithms on wearable or physiological signals (PPG, ECG, accelerometry, sleep or

activity), including owning accuracy against a reference.

Strong study design and statistics. Track record of validation work, with publications or

technical reports.

Fluency in Python and the scientific stack, and comfort reading and reviewing production

code. Experience with ML in production.

Evidence of leading people or technical direction, with a willingness to stay hands-on.

Clear communication of uncertainty to engineers, product leaders and non-technical

stakeholders.

Strong pluses:

Behaviour change or personalised intervention research.

Digital biomarkers or clinical-grade validation experience.

Experience with regulatory pathways for health software and devices.

Experience at a wearables, digital health or medical device company.

Who this is not for:

This role is not a pure research position and not a pure deep learning role. We need someone

who is equally serious about causal reasoning, signal quality and shipping.

What success looks like:

First 90 days: audit current algorithms and accuracy, agree a single validation framework

and release gates, and publish a science roadmap.

First 6 months: measurable accuracy improvements on core metrics, and a first causal-

insight capability running in production.

First 12 months: LifeOS recommendations with evidence of impact on user outcomes, and a

team that runs on shared standards