Staff Backline Engineer – ML/AI

The shape of it

In the posting’s own words

As a Staff Backline Engineer, you will be a technical expert and escalation point for some of the most complex ML/AI issues. You will work across Support, Engineering, Product, and customers to troubleshoot difficult problems, reproduce issues, identify root causes, and drive them to resolution.

What it asks for · 9

• Deep troubleshooting experience with distributed ML/AI systems and the ability to debug problems across application code, frameworks, infrastructure, and the Databricks platform.
• Strong hands-on Python experience and the ability to build, modify, and debug ML workloads using frameworks such as PyTorch, TensorFlow, or Scikit-Learn.
• Strong understanding of Databricks ML/AI technologies, including MLflow, Model Serving, Feature Engineering, Spark MLlib, and model lifecycle management.
• Strong Apache Spark knowledge, including DataFrames, query execution, distributed computing, memory management, shuffles, and performance optimisation.
• Experience troubleshooting training and inference performance, including CPU/GPU utilisation, memory issues, data bottlenecks, concurrency, latency, and distributed execution.
• Experience with ML deployment and infrastructure such as Kubernetes, cloud ML platforms, CI/CD, model monitoring, and production ML systems.
• Ability to read and reason about code, logs, stack traces, metrics, traces, execution plans, and system behaviour rather than relying solely on documentation or configuration recommendations.
• Demonstrated ability to independently own ambiguous, high-impact technical problems and drive them from symptom → investigation → root cause → resolution .
• Strong technical communication skills and the ability to influence Engineering, Product, Support, and customers.

What the job covers

• Serve as a senior escalation point for complex ML/AI issues involving model training, inference, Model Serving, MLflow, Feature Engineering, with knowledge of Spark, Delta Lake, and distributed workloads.
• Perform deep technical investigations using logs, traces, metrics, profiling, configuration, source code, and customer workloads to identify root cause.
• Reproduce customer issues through hands-on experimentation, Python/Spark development, workload construction, configuration changes, and performance analysis.
• Troubleshoot model training and inference failures, performance degradation, resource utilization, memory/CPU/GPU issues, distributed execution problems, and deployment/runtime failures.
• Partner closely with Engineering and Product teams to drive difficult issues to resolution and influence product improvements.
• Identify recurring failure patterns and turn them into better diagnostics, documentation, tooling, automation, and Claude skill capabilities.
• Mentor engineers and raise the technical troubleshooting capabilities of the broader Support organization.
• Act as a technical SME for ML/AI platform areas and contribute to cross-functional initiatives with global impact.

Tools and skills named

Words the posting leans on

• issues9×
• technical9×
• customer8×
• engineering7×
• data6×
• model6×
• platform6×
• distributed5×
• experience5×
• ml/ai5×
• problems5×
• product5×
• deep4×
• drive4×
• execution4×
• infrastructure4×

Counted from the posting after the mission statement and the legal notices are set aside. The ones near the top are the ones a screener is looking for.

More open at Databricks

• デリバリーソリューションアーキテクトTokyo, Japan
• Account ExecutiveBengaluru, India; Mumbai, India
• Account Executive, Commercial - HunterTokyo, Japan
• Account Executive, SingaporeSingapore
• Accounting Data & Technology ManagerMountain View, California; San Francisco, California
• Accounting Manager, Accounts ReceivableAustin, Texas; Denver, Colorado; Portland, Oregon