About Us
Aarki is an AI company that builds advertising solutions to drive mobile revenue growth. We use AI to find audiences in a privacy-first world by using trillions of contextual bidding signals coupled with proprietary behavioral models. Our audience engagement platform includes a full-service agency team and Unified Creative Strategy that delivers ad creative ideation and execution. We have worked with hundreds of advertisers over 14 years and see 5M mobile ad requests per second from over 10B devices driving performance for publishers and brands. It is independently operated and headquartered in San Francisco, CA with offices across the United States, EMEA, and APAC.
Role Overview
Aarki is seeking a results-driven and innovative Data Scientist to join our engineering team. As a Data Scientist for the Aarki team, you will be in the core team building and growing the programmatic demand-side platform (DSP) that leverages advanced machine learning algorithms and state-of-the-art software architecture to deliver performance metrics that beat the industry norms by several magnitudes.
Role & Responsibilities
- Research, design, and implement algorithms and pipelines for solving optimization and machine learning problems, e.g.
- User response prediction, CTR/CVR estimation
- Bid landscape forecasting from censored data
- Bidding and budget pacing strategies
- Fraud and anomaly detection
- Read relevant research and apply learnings to the problem domain
- Analyze system performance relative to various key metrics and establish a framework for fast iterative development
Skills & Experience
- A minimum of 5 years of experience
- Proven track record in one of the following machine learning domains: recommender systems, Bayesian inference, latent variable models, neural networks, deep learning
- Strong working knowledge of machine learning, data modeling, and visualization
- Experience with Python, TensorFlow/Pytorch, ScikitLearn, XGBoost
- Experience with deploying machine learning models to production systems
- Experience with Spark and SQL
- Good understanding of probability and statistics
- Good understanding of numerical optimization algorithms, convergence, and stochastic optimization methods
- Preferred experience with RTB (real-time bidding) or other computational advertising problems
- Preferred experience with Rust, C++, C#, Java, or Scala
- Preferred experience with data warehousing systems and MPP databases
- Preferred working knowledge of auction and game theory
- Preferred familiarity with network and system administration of high throughput/high load environments
- Bachelor's in Mathematics, Physics, Computer Science, or another technical field
- Master’s or Ph.D. preferred
