Eric Zhang

Engineered the Glean AI Assistant and Agents, enabling natural language queries and workflow automation, which saved employees 2-3 hours weekly per user and supported over $100M in ARR growth.

Designed and implemented a robust RAG backend using Python and Java, integrating enterprise search results into LLM prompts to ensure grounded, document-verified answers.

Developed a scalable serverless query path with AWS Lambda and API Gateway, elastically managing Assistant traffic while maintaining low, interactive chat-style latency.

Orchestrated Kubernetes-based microservices with in-memory caching and rate limiting, effectively managing LLM calls and protecting downstream systems as query volume doubled per user daily.

Built the conversational Assistant UI using React and TypeScript, incorporating streaming responses, inline citations, and follow-up prompts to facilitate safe exploration of company knowledge for non-technical users.

Implemented critical policy and safety checks for sensitive data within AI features, ensuring enterprise governance compliance, and mentored junior engineers, elevating backend code quality and accelerating new hire onboarding.

Developed a self-serve ads platform that enabled advertisers to launch and measure campaigns efficiently, significantly scaling Reddit's ads business with year-over-year growth in daily active users and ad spend.

Designed and implemented a Kinesis-to-Lambda-to-S3 data pipeline in Python, ingesting impression and click events to provide reliable raw data for billing, optimization, and downstream analytics.

Built responsive advertiser and internal tools using React and Angular, translating UX wireframes into accessible interfaces and validating flows with Selenium, Cucumber, and JUnit automation.

Extended Java-based creative services to support image/video uploads to S3 and metadata APIs, enhancing ad formats, and implemented Spring Boot/Java targeting and bidding services with PostgreSQL, Redis, and ActiveMQ for traffic management.

Managed robust CI/CD workflows with GitHub and Jenkins for microservices and frontend applications, ensuring automated deployments to AWS ECS/EKS, and utilized Grafana, New Relic, Splunk, and CloudWatch for monitoring and capacity planning.

Developed and optimized core streaming and discovery services, enabling millions of members to instantly stream content without interruption across a 100M+ global subscriber base.

Automated maintenance and incident response with Python, SQL, and Bash scripts on Linux, significantly increasing debugging speed and reducing manual operational workload for the on-call team.

Authored comprehensive JUnit, TestNG, and JMeter performance test suites on Linux, rigorously validating API correctness, load behavior, and regression impact for large-scale rollouts.

Engineered robust CI/CD pipelines utilizing GitHub, Jenkins, Docker, and Kubernetes, automating testing, container builds, and multi-region deployments for critical edge and metadata services.