Justin Pachter

I’m a Senior Software Engineer with 13 years of experience building high-scale distributed systems and AI-integrated platforms. My focus is turning data-heavy problems into low-latency, reliable services—especially where accuracy, safety, and real-time decisioning matter.

At Google, I led end-to-end engineering for financial security across Google Pay and content safety for YouTube. I architected high-throughput Java and Python services that reduced fraud losses by 30% and improved moderation accuracy by 50% for billions of users.

I build and operationalize systems that perform under pressure: transaction microservices for concurrent payment streams, real-time anomaly detection integrated with Vertex AI and TensorFlow, and performance targets like under 150ms during payment authorization. I also optimize persistence and caching with Cloud Spanner and Redis to maintain sub-100ms P99 response times, while running stable service boundaries and rigorous load testing.

I’ve also driven AI and platform capabilities beyond the backend—RAG frameworks to ground moderation decisions in up-to-date legal policy datasets, tiered filtering to cut Gemini and PaLM 2 API compute costs by 40%, and moderator-facing interfaces that make human review efficient. Earlier, at eBay, I scaled a real-time personalization engine to increase cross-sell conversion by 25%, stream processing with Kafka/Spark, and deployment automation that reduced delivery cycles from two days to under one hour.