Senior Data Scientist

AtmosphereTV; Austin (TX); On-site; USD 130,000 - 190,000; Full time

Benefits offered by this job

Company Equity

401(k) with employer matching

Competitive insurance plans

Flexible Time Off Policy

Job summary

Atmosphere TV is hiring its first data scientist to build the causal-inference and predictive-modeling foundation for measuring and optimizing advertising and venue growth.

You will design frameworks, run experiments, and develop models to predict venue revenue, retention, and campaign performance, partnering with product, sales, and marketing in a fast-moving, data-driven environment.

Qualifications

• 5+ years of data science, applied research, or quantitative analytics.
• Deep fluency in causal inference and applied statistics.
• A/B testing, geo experiments, difference-in-differences, synthetic control.
• Expertise in predictive modeling (gradient-boosted trees, churn models).
• Strong Python and/or R skills; SQL for large-scale data.
• Ability to translate statistical work into clear business outputs and ship data products.

Responsibilities

• Design and own Atmosphere's causal measurement framework—isolating incremental impact from confounders.
• Build statistically rigorous designs: exposed/control groups, geo experiments, diff-in-diff, synthetic control.
• Stand up the incrementality capability to support sales and GTM strategies.
• Build models predicting a venue's intrinsic revenue potential and retention drivers.
• Leverage venue streaming data to understand churn and prioritize improvements.
• Sharpen audience segments and model features for revenue and retention.
• Develop closed-loop optimization frameworks and proprietary benchmarks.

Skills

Data science

Causal inference

Statistics

A/B testing

Geo experiments

Difference-in-differences

Synthetic control

Predictive modeling

Python

R

SQL

Model evaluation

Education

Advanced degree or相关 équivalent in a quantitative field

Tools

Python

R

SQL

Job description

Atmosphere TV is the leading streaming TV platform built specifically for businesses. Unlike ad networks or signage companies, Atmosphere is the only true TV company whose first priority is to entertain television audiences outside of the home. Our content is designed to be fun, engaging, and worth watching, transforming waiting rooms, gyms, bars, and restaurants into better experiences for customers and better businesses for owners.

We are the first and only company to think about both the business owner and their customers when creating TV content. With 60,000+ venues and a global audience of over 150 million monthly viewers, Atmosphere TV is redefining what TV means outside the living room.

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.

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.
• 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.
• 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.
• 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.