About the role
As we scale both our advertising business and our venue network, we are investing in the causal-inference and predictive-modeling foundation that will make Atmosphere smarter on both sides of the business — how we measure and sell advertising, and how we grow and retain the venues that make up our network. As our first data scientist, you will help shape this function from the ground up.
We are looking for a Senior Data Scientist to serve in a high-impact, high-visibility role that sits at the intersection of data science, sales, product, and go-to-market. The ideal candidate is an excellent applied statistician and modeler: someone who deeply understands causal inference and predictive methods, knows which technique fits which problem and why, and can turn that rigor into products the business actually uses. You will design, build, and ship the models that quantify the real-world impact of campaigns, capture what makes individual venues valuable and what puts them at risk, and turn rigorous methodology into a core competitive advantage for Atmosphere.
What you'll do
Causal Measurement & Incrementality
- Design and own Atmosphere's causal measurement framework — isolating the incremental impact of exposure on real-world outcomes (foot traffic, store visitation, conversion) from confounders like organic visitation trends, seasonality, and competing media.
- Build statistically rigorous causal designs: exposed/control group construction, geo-based experiments, difference-in-differences, and synthetic control.
- Stand up the incrementality capability that underpins how we sell — turning measurement into a differentiator our go-to-market teams can take to market and our clients can trust.
Predictive Venue Modeling & Contextual Enrichment
- Build models that predict a venue's intrinsic revenue potential, enabling us to prioritize prospective venues for acquisition as well as flag under-monetized venues we already operate.
- Leverage venue streaming and engagement data to better understand what drives retention vs. churn, helping us create a ranked list of customer challenges to solve from the full list of potential friction points, pain signals, and leading indicators.
- Model and infer latent venue attributes to sharpen both the segments advertisers target against and the feature sets your own revenue and churn models depend on.
Productization & Go-To-Market
- Use measurement outputs to generate actionable insights that feed back into campaign strategy — daypart targeting, venue type optimization, creative performance, audience segmentation
- Develop closed-loop optimization frameworks that make every campaign smarter than the last, building proprietary benchmarks and norms by vertical/category
Qualifications
- 5+ years of experience in data science, applied research, or quantitative analytics.
- Deep fluency in causal inference and applied statistics – including experience with A/B testing, geo experiments, difference-in-differences, synthetic control, propensity methods, and regression modeling – and can explain clearly why one fits a given problem better than another.
- Expertise in predictive modeling (e.g., gradient-boosted trees, survival/churn models, calibration, and honest out-of-sample evaluation)
- Sound judgment about what makes a model trustworthy: validation, uncertainty, and knowing when a result is solid enough to act on.
- Strong programming skills in Python and/or R, and comfort with SQL for working with data at scale.
- A track record of translating complex statistical work into clear, business-friendly outputs, and of shipping models and products that get used
- Ability to operate in a fast moving environment and balance rigor with pragmatism.
Nice-to-Have:
- Familiarity with location/mobility data or other behavioral signal data.
- Familiarity with out-of-home (OOH), digital out-of-home (DOOH), or CTV advertising, or experience at an ad tech company, media platform, or measurement vendor.
- Exposure to marketing measurement, media effectiveness, mixed media modeling (MMM), or multi-touch attribution (MTA).
- Background in Bayesian inference or probabilistic modeling.
- Experience launching data products in coordination with Sales, Marketing, or Client Success.
- Exposure to computer vision or NLP..
Compensation & Benefits:
- Competitive salary
- Company Equity
- Company 401(k) with employer matching
- Competitive insurance plans
- Flexible Time Off Policy
Our Commitment to Diversity:
Don’t meet every single requirement? Research shows that women and underrepresented groups often hesitate to apply unless they meet all the criteria. At Atmosphere, we’re committed to building a diverse, inclusive team where creativity, innovation, and teamwork thrive. If you're excited about this role but your experience doesn’t perfectly align with every qualification, we still encourage you to apply—you might be the right fit for this or another role.