Cloud and Telemetry team-Building and managing infrastructure for firm-wide telemetry. Building a hot-hot architecture resilient to full regional failures.- Leading modernization of firm-wide telemetry.- Working with OS teams to set up firm-wide telemetry agent.- Managing and developing self-service workflows and infrastructure for telemetry tasks- Managing and developing firm-wide telemetry forwarding and enrichment layer- Training the team on telemetry day-to-day operations.- Writing improvements to Open Telemetry Collector and other telemetry processors.- Finops, cost controls and cost allocations for Cloud and Telemetry products

A community member, mentoring new participants, and organizing the project.- Running Beam Summit 2021 (https://2021.beamsummit.org)- Reviewing code, participating in technical and community discussions- Discussing committership for community members among committee members

Contributing to Apache Beam's Python and Java SDKs all across the stack (IO connectors, internal instrumentation, pipelines, etc).Also spending time helping others contribute via code reviews, intern and GSoC student mentoring, answering questions on StackOverflow and mailing lists.

Working all across the stack on Apache Beam and Google Cloud Dataflow. - Developing an industry-standard open source library for batch and low-latency stream processing (Apache Beam).- Educating others and building community around it (sample presentation https://s.apache.org/the-dataflow-sausage)- Developing data pipelines (IO connectors, SQL-based transformations)- Developing the data processing runtime (performance-critical data movement, distributed task management).Since 2022Leading development of a low-code product for real-time analytics based on Beam + Dataflow.- Developing a high-performance pipeline builder, with data semantic features and SQL (https://github.com/GoogleCloudPlatform/DataflowTemplates/tree/main/syndeo-template)Since 2019Leading CDC features on Beam and Dataflow- Prototyped Debezium-based connectors (https://github.com/pabloem/beam/tree/debeziumio-prototype/sdks/java/io/cdc), and led implementers (https://github.com/apache/beam/pull/13983)- Defining semantics for per-key ordering in Beam - to enable users and customers to deploy their use cases confidently (https://github.com/apache/beam/pull/15378)Interesting technical links: - Designing a new type-safe API for key range partitioning: https://github.com/apache/beam/pull/15848 - Designing a new testing API for Airflow DAGs: https://docs.google.com/document/d/1priak1uiJTXP1F9K5B8XS8qmeRbJ8trYLvE4k2aBY5c/edit#heading=h.bpa8abweu50h

Developing a distributed task manager in Java.

Designing a semantic database in Node.js that updates itself automatically. Required parallel programming, caching, probabilistic decision making and probabilistic data structures. Tons of fun.

Interning in the Knowledge Engine team. Developing features that use Google's semantic database/graph.

Working on a predictive model to estimate probability of failure of software tests, to minimize the number of test suites run in an automatic testing platform.

SDE in the Oracle Database Scheduler team.