Software engineer, connectors & MCP

present of AI; San Francisco (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Writer is seeking a backend engineer to build the plumbing that lets AI agents reach outside systems. You will work on MCP clients and servers and the APIs that carry agent workflows, using TypeScript/Node stacks and implementing OAuth and token handling for third-party integrations.

Reliability and observability are as important as feature delivery. Join a team focused on secure, observable, and dependable agent actions at scale.

Qualifications

• Production backend experience in TypeScript/Node-style stacks.
• Experience with auth patterns like OAuth and token handling for third-party integrations.
• Reliability work: metrics, service level targets, incident readiness.

Responsibilities

• Build backend plumbing: connectors, MCP clients/servers, and APIs for agent workflows.
• Implement secure, observable integrations between AI agents and outside systems.
• Design and maintain scalable, reliable backend services with proper failure handling.

Skills

TypeScript

Node.js backend

OAuth & token handling

Observability & metrics

Reliability & SLOs

Job description

This role builds the backend plumbing that lets AI agents reach outside systems: connectors, MCP clients and servers, and the APIs that carry agent workflows. A strong candidate has years of production backend work in TypeScript or Node style stacks, plus real experience with auth patterns like OAuth and token handling for third party integrations. Reliability work such as metrics, service level targets and incident readiness matters as much as feature delivery.

How it fits the AI era

The work sits at the seam between AI agents and the outside world, so much of it is about making agent actions secure, observable and dependable. AI can speed up scaffolding and routine code, but trust boundaries, auth design and failure handling still need human judgment. The role grows as more enterprise systems get connected to agents.

How it builds AI experience

A person here would gain hands on practice building MCP clients and servers and tool orchestration layers, which is directly the kind of integration work the company sells. That means learning how agents call external tools safely, how identity is mediated, and how to keep such calls reliable at scale. Prior prompt or model tinkering alone is not the same as this integration depth.

Worth showing

• MCP client and server work
• OAuth2 and OIDC flows
• TypeScript backend services
• API and distributed systems design
• SLOs and observability practice

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

About Writer

Writer is an enterprise AI platform for building and supervising AI agents grounded in company data, per the company.