Leading ML Monitoring efforts for Moloco's advanced AI/ML models that power programmatic advertising at scale.

• Technical lead for ML Monitoring project to develop tools for end-to-end model diagnosability and training debugging applied for Meta's AI models and training pipelines to debug feature, QPS, and model issues.

• Technical lead for Efficient Investigations project to develop automated tools for runbooks used for debugging production issues across Meta.

• Technical Lead and founder for Dataflow Tracing project to create tools for tracing data flows and its analysis to debug issues such as data loss and latency for large scale dataflow systems.

• Technical Lead for the service management layer of Meta's stream processing system.

• Storage startup backed by LightSpeed, Menlo, Kleiner Perkins

• Responsible for end-to-end design and implementation of the entire distributed monitoring and analytics stack for a scale out storage array.

• Additional contributions made to PernixData FVP product (a distributed scale-out server-side cache).

• Early stage Industrial IoT startup that provides software to manage industrial sensor data

• Providing technology consulting and advice

• Worked on and led projects on management of big data systems

• Topics include scalable monitoring for troubleshooting multi-stage big data systems, resource management for big data applications on CPU-GPU systems, cluster scheduling improvements for big data workloads

• Technical lead and managed a team of researchers and interns