Principal Architect

We are looking for someone who can look at a distributed system under load and know, within minutes, whether it's about to fail gracefully or catastrophically, and who has already spent a career building the muscle to make it fail gracefully. Someone who has shipped code themselves for long enough to distrust architecture that only exists on a whiteboard and who now wants to point that instinct at one of the hardest problems in applied AI: making multi-agent, RAG-driven systems behave predictably in production.

Responsibilities:

• Architect distributed systems and developer infrastructure that hold up under real production load,
• Prototype hands-on to prove an architecture before it becomes a standard others build against.
• Participate actively in code review, catching failure modes and design issues others may miss.
• Turn architecture decisions into technical documentation clear enough that another senior engineer could challenge the reasoning without a meeting.
• Own the tradeoffs between performance, cost, and reliability, and defend those tradeoffs with evidence.
• Define the organizational tech stack and coordinate technical direction across teams, ensuring there is no unmanaged technology sprawl.

Requirements:

• Demonstrated experience architecting high-performance, distributed systems and developer infrastructure that have run in production.
• Full-stack engineering depth: able to reason about and personally build across the stack, not just direct others who do.
• A track record of building systems that are trustworthy and cost-efficient under real failure conditions, with evidence to back it.
• Experience designing end-to-end multi-agentic GenAI/RAG pipelines for ingestion, chunking, vector search, message passing, and orchestrator patterns that stay reliable as complexity grows.
• Strong command of design patterns, state management, and failure-handling patterns for distributed and agentic systems.
• Experience architecting multi-agentic workflows and end-to-end GenAI/RAG pipelines: ingestion, chunking, vector search, message passing, and orchestrator patterns.
• An appetite for solving hard, ambiguous problems in distributed and agentic system design, where the right answer isn't established yet.

Tech Stack:

• Backend and Infrastructure: Python (FastAPI, async/await, Pydantic), PostgreSQL (schema design, indexing, migrations), Kafka / Google Pub/Sub, Docker, Kubernetes (GKE), GCP (Cloud SQL, GCS, GKE), multi-cloud patterns, Prometheus and structured logging, OpenAPI/Swagger, API versioning, auth patterns.
• Frontend: Next.js, React, TypeScript, JavaScript, REST APIs, TanStack Query, SSR / SSG / CSR / ISR, HTML5 CSS3 framework-agnostic UI development, frontend performance and bundle optimization, testing practices, CI/CD pipelines.
• AI and Distributed Systems: Distributed Systems Design, Microservices and Service Orchestration, Vector Search and Embeddings, RAG Pipelines (Ingestion, Chunking, Retrieval), Multi-Agent Orchestration Frameworks, Message Queues / Event-Driven Architecture, Performance Profiling and Cost Optimization, Data Intelligence and Curation, Distributed AI Infrastructure.

Good to Have:

• Experience standing up developer infrastructure or platform tooling used by other engineering teams.
• Exposure to on-device or resource-constrained AI deployment.
• Background contributing to or reviewing architecture for early-stage, fast-moving engineering organizations.
• Formal chaos engineering or failure-injection testing experience.