Growth Data Scientist

Kalshi•New York City, NY

5d•$200,000 - $250,000

About The Position

Kalshi is defining a new category Kalshi has defined a new category: prediction markets. Kalshi allows people to trade on the outcome of any events and turn any question about the future into a financial asset. Kalshi fought for years and legalized prediction markets in the US for the first time in history, is currently the fastest growing financial market in America, and has thousands of markets across politics, economics, financials, weather, tech, AI, culture and more. We believe prediction markets have the potential to be the largest financial market because they turn anything into a financial position. Our vision: well… build the largest financial market on the planet. Our mission: bring more truth to the world through the power of markets. Our culture is simple: we hire really talented people, work really hard, and enjoy the climb. We are looking for ambitious and exceptional people to join our (relatively small) team to help us build the next generation of financial markets. Role Roadmap Kalshi is seeking a Senior Data Scientist to own and define the company’s Customer Lifetime Value (LTV) framework as a core strategic input to growth and product decisions. This individual will ensure LTV reflects durable, structural drivers of customer value - not short-term seasonality or promotional noise. They will serve as Kalshi’s authority on long-term value, building models that generalize across geographies and regulatory environments, and translating nuanced analysis into clear guidance for leadership as the company scales.

Requirements

5–8+ years of experience as a Data Scientist in a consumer marketplace with meaningful seasonality
Proven ownership of an LTV model that:
Lived in production for multiple years
Survived multiple seasonal cycles
Informed real budget, growth, or product decisions
Deep experience with:
Cohort-based modeling and survival analysis
Feature-rich segmentation (geo, behavior, product mix)
De-biasing early lifecycle and promotion-heavy data
Strong SQL and Python (or R); comfortable working with large-scale feature pipelines
Demonstrated ability to explain why a model generalizes—not just that it performs

Responsibilities

Design LTV models that capture structural—not just seasonal—value
Separate long-term signal from short-term promotions and seasonality
Model retention and monetization dynamics that persist across cycles
Stress-test models across cohorts launched in different macro and regulatory regimes
Incorporate high-granularity features
Build user- and cohort-level models that account for:
Geography (e.g., zip code, DMA, state-level effects)
Regulatory or market-specific constraints
Behavioral and product-usage signals
Explicitly model heterogeneity rather than relying on global averages
Ensure durability and generalization
Validate models across multiple years, product launches, and seasonal cycles
Monitor stability, drift, and cohort aging effects
Revisit assumptions as the business and user base evolve
Make LTV actionable at different resolutions (user, cohort, geo, channel)
Partner with Growth to avoid overfitting CAC decisions to short-term spikes
Align with Finance on long-range forecasting assumptions
Be the voice of judgment
Push back against simplistic or purely seasonal interpretations of value
Clearly communicate uncertainty, confidence intervals, and limitations
Prevent misuse of early-cohort or promotion-inflated signals