[Remote] Lead Platform Engineer, API & AI Services
Note: The job is a remote job and is open to candidates in USA. Apolis is a company that focuses on delivering AI solutions for legal research and document management. They are seeking a Lead Platform Engineer to own and secure the production API gateway for their CoCounsel platform, while leading a team of engineers and collaborating on AI service integrations.
Responsibilities
- Own and operate the production multi-tenant API gateway: CIAM/OIDC authentication, per-tenant routing, rate-limiting, fault isolation, and rolling deploys on AWS ECS Fargate. Maintain and evolve JWT validation — JWKS key management, claim validation, tenant identity propagation, and key rotation. Drive reliability through circuit-breaking, latency budgets, and graceful degradation
- Own the CIAM/OIDC/OAuth 2.0 integration at the gateway layer — this is the primary security perimeter for enterprise customer traffic. Implement and maintain JWT validation, tenant-scoped authorization, and API key lifecycle management
- Lead a pod of 3–4 engineers: code reviews, architectural decisions, delegating concrete pieces of work with clear ownership, and mentoring. Resolve cross-team design questions peer-to-peer. Hold a clear distinction between DEV iteration (fast, reversible, learn by breaking) and PROD discipline (coordinated, observable, controlled rollout)
- Maintain CI/CD pipelines and IaC (CDK or Terraform) for the gateway and its supporting infrastructure. Own AWS infrastructure — ECS Fargate task lifecycle, multi-region rollout, ElastiCache Redis for distributed state, and RDS for quota and analytics. Build and maintain observability: distributed tracing, latency dashboards, per-tenant usage metrics
- Participate in the design and engineering of LLM-backed sub-services the team builds and operates: contract analysis pipelines, RAG architectures, agentic workflows. Contribute to decisions about how AI service results propagate across the platform — async job patterns, structured outputs, webhook delivery — and bring a platform engineer’s perspective to reliability, observability, and integration
- Define and build how enterprise customers access CoCounsel’s capabilities beyond REST APIs. Enterprise customers are already asking for MCP servers, callable tools, and programmatic interfaces their own AI agents can invoke. This is an open design problem on a near-term roadmap — you will shape the answer, building directly on the gateway and auth infrastructure you own
Skills
- Python/FastAPI in production
- You have built and operated a FastAPI service handling real external traffic — not just internal tooling or AI-app backends
- AWS
- Hands-on with ECS Fargate, CloudFormation
Benefits
- Only W2
Company Overview