Christopher Song

About

Full-stack engineer who built and owned a production real-time computer-vision inference platform at Credence, spanning Next.js App Router/TypeScript frontend with SSE/WebSocket streaming, a Flask backend, and Postgres analytics. Demonstrated measurable performance wins (70% fewer re-renders; latency cut to ~40–50ms) and strong production rigor (durable orchestration, idempotency, observability, AWS EC2 + CI/CD) with tight post-launch UX iteration based on analyst feedback.

Key Strengths

• Shipped end-to-end real-time vision monitoring feature (Next.js App Router + TS + Flask) with live detection/segmentation overlays
• Designed hybrid data-fetching strategy (server-first initial load + client-side streaming) for low-latency real-time UX
• Resolved Next.js App Router caching/staleness issues using no-store and dynamic route configs for fresh session state
• Optimized high-frequency UI rendering by moving frame updates to canvas/useRef; reduced re-renders ~70% and improved interaction latency from ~120–150ms to ~40–50ms
• Postgres schema and query optimization at scale (100k+ detections); reduced dashboard query latency ~800ms to ~90–120ms using indexing, query rewrite, and materialized views
• Built durable, idempotent workflow orchestration for multi-step inference pipelines with retries, checkpoints, and reconciliation to prevent duplication/data loss
• Strong production ownership: AWS deployment, CI/CD, observability, performance tuning, and iterative UX improvements driven by instrumented user feedback

Experience

Education

Awards

• Dean’s List - University of Maryland, College Park (Fall 2023)
• Dean’s List - University of Maryland, College Park (Spring 2024)

Skills

Languages