We are building a high-performance, distributed AI orchestration platform that processes complex, multi-step pipelines at scale. This is not a typical SaaS role; you will own architecture decisions, system design, and delivery velocity. We ship one production-grade feature every 2 weeks. You will use AI-assisted development tools (Cursor, Windsurf, Claude Code, or Antigravity) daily as your core workflow, not as a novelty. These tools are how we achieve senior-engineer velocity without sacrificing quality. The core responsibilities for the job include the following: Expert-level distributed systems knowledge you will own: • Async job orchestration (Temporal.io ) and event-driven pipelines. • Multi-service coordination at scale with service-oriented architecture. • Eventual consistency, the CAP theorem, and scaling patterns (sharding, partitioning, and replication). • Production Redis experience with pub/sub, caching, and cluster management. Advanced Postgres and Supabase (production level): • Complex schema design, query optimization, indexing. • Row-level security, triggers, and stored procedures. • Multi-tenant data isolation, migrations at scale. AI pipeline reliability design systems with: • Idempotency, resumable workflows, retry strategies. • Cost tracking and spend guardrails. • Clear logs and error states. • Full-stack capability owns features end-to-end (schema API UI deployment), but with deep backend expertise. You spend 70% of your time on backend/infrastructure and 30% on frontend. System design and architectural thinking: • Architect solutions for 10M+ events/day without hand-holding. • Design with clarity, identifying trade-offs before implementation. • Think in terms of constraints, dependencies, and long-term maintainability. Spec-driven development: • Read specs thoroughly and flag ambiguities early. • Design APIs and schemas that match the spec intent. • Use specs to guide implementation and AI generation. Production-grade AI-assisted development (must have real shipping experience): • Deep mastery of at least one: Cursor, Windsurf, Claude Code, or Antigravity. • Expert context management structuring prompts, codebase context, and specs for AI tools. • Shipped production features using AI tools without sacrificing quality or introducing regressions. • The reality: Excellent engineers with AI ship faster and better. Poor engineers with AI introduce more bugs. Senior engineering mindset: • Think about trade-offs, not just make it work. • Flag architectural risks before they become production incidents. • Mentor through code and design. The core requirements for the job include the following: Strong advantages: • Experience of server-side video generation using AI models (Veo, Kling, Hailu, Seedance, Sora, Wan, etc. ). • Experience of concurrent video/image generation using heavy AI models. • Strong knowledge of video generation concepts (text-to-video, image-to-video, and image-to-video with reference images). • Experience of image generation using AI models (NanoBanana, Wan, etc. ). • Google Vertex AI experience. • < 1 week feature shipping cadence. • UGC domain knowledge workflows, asset pipelines, AI-assisted tools (Higgsfield, Speel), video generation. Tech stack proficiency: • Runtime: Node.js . • Frontend: Next.js (App Router), TypeScript (strict mode). • Backend: NestJS, TypeScript (strict mode). • Cloud: GCP (Vertex AI), AWS (EC2 EKS, RDS, Lambda). • Deployment: Vercel, Docker, Kubernetes. • CI/CD: GitHub Actions. • AI Models: Claude, Gemini, Vertex AI, Veo, Kling, Hailu, Seedance. • Testing: Playwright.