About this role
AppLovin builds end-to-end advertising solutions that help businesses connect with global audiences across a platform reaching over one billion users. Software engineers on this team advance user signal and recommendation technologies by working on large-scale machine learning problems that span ranking, retrieval, model architecture, and optimization.
This role exists to improve the relevance and performance of advertising systems by developing new ways to represent and utilize user signals. Professionals joining this group will manage the entire machine learning lifecycle, from feature development and modeling to experimentation and production deployment. Ideal candidates bring solid programming abilities, deep learning experience, and a background in scaling production machine learning models.
What you will do
- Develop and improve user signals, features, and representations used by large-scale machine learning models for advertising and recommendation.
- Explore machine learning approaches to learn effectively from large-scale, sparse, noisy, and heterogeneous user signals.
- Advance large-scale recommendation systems across candidate retrieval, ranking, prediction, and optimization.
- Scale machine learning models and training systems to support increasing data volume, model complexity, and computational requirements.
- Improve training and inference efficiency by identifying bottlenecks across model computation, data loading, memory utilization, distributed execution, and hardware utilization.
- Design and analyze offline and online experiments to understand the incremental value of user signals and model improvements.
- Collaborate closely with engineering, data, and product teams to bring new user signals and machine learning approaches from experimentation into production.
What you need
- Hold a Bachelor's degree in Computer Science, Computer Engineering, Machine Learning, or a related technical field, or possess equivalent practical experience.
- Bring at least 4 years of experience developing and deploying machine learning systems in production environments.
- Demonstrate experience with machine learning or deep learning in areas such as recommendation, ranking, retrieval, prediction, advertising, or representation learning.
- Show hands-on experience developing and training machine learning models using large-scale datasets.
- Possess a strong understanding of machine learning fundamentals, including model architectures, optimization, representation learning, feature engineering, and model evaluation.
- Exhibit strong programming and software engineering skills with a track record of building reliable production systems.
- Bring experience using modern deep learning frameworks such as PyTorch or TensorFlow.
Nice to have
- Experience developing user signals, features, or learned user representations for large-scale machine learning systems.
- Experience with distributed model training and large-scale ML infrastructure.
- Experience optimizing training or inference workloads on GPUs or other accelerators.
- Experience optimizing ML systems for latency, throughput, memory utilization, or computational efficiency.
Benefits
- Equity eligible
- Health Insurance: Medical, Dental, Vision, Life, Disability
- 401(k) Retirement Plan
- Unlimited Discretionary Time Off
- 10 paid holidays per year
- 80 hours of paid sick leave per year
Skills and tech stack
Questions candidates ask
What is the salary range for this role?
The California base pay range is $221,000 to $331,000 USD. Total compensation is based on factors including market location and may include equity and other forms of incentive compensation.
What kind of experience is required?
Candidates need a relevant Bachelor's degree or equivalent practical experience, along with 4+ years of experience developing and deploying machine learning systems in production environments. Strong programming skills and familiarity with frameworks like PyTorch or TensorFlow are also required.
What will I be working on day-to-day?
You will work on large-scale machine learning problems spanning user signals, representation learning, ranking, retrieval, model architecture, and optimization. Your responsibilities include moving models from feature development and experimentation into production.
How does the hiring process work?
The application window is expected to close within 30 days of the posting date. To support an efficient and fair hiring process, technology-assisted tools including AI may be used to identify and evaluate candidates, though all final hiring decisions are made by human reviewers.