Snap
Software Engineer, ML Infrastructure, ML Platform Level 5
$178K–$313K
Los Angeles, California +2 · Full-time · Posted Oct 15, 2026
- Location
- Los Angeles, California
Palo Alto, California
Bellevue, Washington - Experience
- 6+ years of post-Bachelor’s software development experience; or Master’s degree + 5+ years of post-grad experience; or PhD in a relevant technical field + 2+ years of post-grad experience.
- Education
- Bachelor’s degree in a technical field such as computer science or equivalent experience.
About the job
Build and operate scalable ML infrastructure and platform experiences for training, serving, evaluating, and managing foundational models at Snapchat.
What you'll do
- Design and improve reliable, efficient infrastructure for large-scale ML workloads, including feature pipelines, inference, training, and evaluation.
- Build data management systems for collection, labeling, processing, and evaluation, and deploy models to production with ML engineers.
- Use AI tools and high-velocity workflows to design and ship secure, scalable services with production-quality code.
What they ask for
- Strong programming skills in Python and Java.
- Experience building large-scale production machine-learning systems, distributed systems, or big-data processing systems.
- Understanding of distributed systems and infrastructure for large-scale ML; experience operating highly available systems at significant scale.
- Experience with big-data processing frameworks such as Spark, Flink, or Ray.
Nice to have
- Experience with ML training platforms or optimizing AI model inference.
- Familiarity with TensorFlow, PyTorch, Caffe2, Spark ML, scikit-learn, or related frameworks.
Pay
$209,000-$313,000 annually (USD, base salary); $199,000-$297,000 annually (USD, base salary); $178,000-$266,000 annually (USD, base salary)
Tools
Python, Java, Go, ADK, langfuse, Spark, Flink, Ray
AI focus
Build agentic experiences and foundational infrastructure for ML models, including model orchestration and lineage, training, evaluation, data quality, and high-performance inference and serving.
Model-written summary · the original posting is authoritative