Staff / Sr. ML Engineer, Indexing Infrastructure

Apple • Santa Clara, CA

5d

About The Position

Join Apple’s MLPT & Infrastructure teams to build the storage and compute foundation for the universal search engine used across products including Siri, Spotlight, Safari, Messages, and Lookup. The role focuses on systems that support search and Apple Intelligence experiences at trillion-record and petabyte scale.

Design and build a web signal store using Parquet and Iceberg, along with an in-house MapReduce-style processing framework that runs thousands of workers on Kubernetes. Improve performance, reliability, and cost across large-scale data storage and processing, and take ownership of production health through monitoring, on-call work, root-cause analysis, and engineering reviews.

Requirements

• Strong coding skills and a solid computer science background in algorithms, data structures, and distributed systems.
• At least 5 years of experience designing, building, and operating large-scale distributed systems such as databases, storage platforms, or data processing frameworks.
• Hands-on experience with petabyte-scale batch data processing using MapReduce, Spark, or similar technologies.
• Proficiency in cloud-native architectures and compute orchestration on AWS (S3, EC2, EKS) and Kubernetes.
• Excellent interpersonal skills and the ability to work independently and drive cross-team efforts.
• MS or PhD in Computer Science or a related field.

Nice To Haves

• Deep familiarity with columnar and lakehouse formats such as Parquet and Iceberg, and the judgment to choose between them at scale.
• Experience with storage internals, including compaction, metadata and manifest management, indexing, and query performance tuning.
• Experience with workflow orchestration (e.g., Argo) and resource scheduling on shared Kubernetes clusters.
• Experience with key-value or online serving stores (e.g., Bigtable, HBase, Cassandra) and multi-cloud data movement.
• Experience with search indexing, machine learning, or NLP.

Responsibilities

• Design and build a petabyte-scale Parquet/Iceberg lakehouse for web documents and signals.
• Design and build an in-house MapReduce-style processing framework running thousands of workers on Kubernetes.
• Improve performance, reliability, and cost at trillion-row scale, including read/write throughput, compaction, and fault-tolerant batch processing on shared clusters.
• Own production health, including monitoring, on-call responsibilities, and root-cause fixes.
• Raise the engineering bar through design and code reviews.
• Drive cross-team efforts and work independently.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior