Skip to main content
TALENT247 // SPECIALIST AI EXECUTIVE SEARCH & TECHNICAL RECRUITMENT
APP · Applied AI

Senior Backend Engineer, Agent Connectors & MCP Platform

Seattle, WA · Full-time · Hybrid

AI Specialism

Applied AI

Workplace

Hybrid

Location

Seattle, WA

Compensation

$190k – $270k

annual package

Role Dossier & Specification

REF: SENIOR-BACKEND-ENG

THE CLIENT & MANDATE

Our client is a premier enterprise work-AI and knowledge platform scaling rapidly across Palo Alto and Seattle. As enterprise adoption evolves from read-only search toward autonomous cross-application action execution, their Agent Connectors & Tooling Platform team builds the secure execution fabric enabling AI agents to read and write safely across Salesforce, Jira, GitHub, Slack, ServiceNow, Workday, and custom internal APIs.

They have partnered with Talent247 to hire a Senior Backend Engineer for their Agent Connectors & MCP Platform in Seattle, WA. The successful candidate will architect universal Model Context Protocol (MCP) gateways, per-user OAuth2/OIDC token delegation brokers, and transactional human-approval guardrails for enterprise agent fleets.

KEY RESPONSIBILITIES

  • Architect a high-throughput, multi-tenant Model Context Protocol (MCP) server and connector gateway exposing hundreds of third-party SaaS and internal enterprise APIs as typed, LLM-ergonomic tool schemas.
  • Engineer fine-grained delegated identity brokers (OAuth 2.1, OIDC, SAML, scoped on-behalf-of token exchange) ensuring agents strictly inherit the invoking human user's live permissions on every read and write call.
  • Build declarative policy engines, dry-run diff previews, idempotency keys, and asynchronous human-in-the-loop approval gates before agents execute mutating external actions.
  • Design adaptive per-tenant rate-limiters, circuit breakers, webhook/CDC incremental sync pipelines, and schema-drift detectors resilient to upstream SaaS API changes.
  • Collaborate with applied AI engineers to optimize tool descriptions, error-recovery hints, and pagination semantics so models achieve 99%+ first-try tool-call validity.
  • Deliver enterprise-grade SDKs and self-serve connector builders allowing Fortune 500 platform teams to register proprietary internal microservices in minutes.

TECHNICAL ENVIRONMENT & OFFERING

Stack: Go, TypeScript, Python, gRPC, Model Context Protocol (MCP), OAuth 2.1 / OIDC / KMS, PostgreSQL, Redis, Kafka/PubSub, Kubernetes, GCP/AWS. Workplace: Hybrid in Seattle, WA. Compensation: $190,000 – $270,000 base salary plus generous late-stage equity and full US benefits.

Core Requirements

ESSENTIAL QUALIFICATIONS

  • 5+ years of backend and distributed systems engineering experience building API platforms, integration fabrics, identity/security systems, or cloud infrastructure in Go, TypeScript, Java, or Rust.
  • Deep practical mastery of OAuth 2.0/2.1, OIDC, webhook architectures, distributed rate limiting, idempotency, and multi-tenant secret management.
  • Experience designing clean, strongly typed API schemas (OpenAPI, JSON Schema, Protobuf, or MCP) consumed by external developers or AI agents.
  • Proven track record operating high-reliability services on Kubernetes with comprehensive structured logging, distributed tracing, and tenant isolation.
  • Security-first engineering mindset accustomed to enterprise threat modeling (confused-deputy prevention, SSRF mitigation, audit logging).

PREFERRED / BONUS EXPERIENCE

  • Direct hands-on experience building or hosting production Model Context Protocol (MCP) servers, agent tool sandboxes, or iPaaS connector engines.
  • Familiarity with how LLMs parse tool schemas, handle structured error recovery, and remain vulnerable to indirect prompt injection via untrusted tool outputs.
  • Prior experience integrating complex enterprise SaaS APIs (Atlassian, Salesforce, Microsoft Graph, GitHub, Google Workspace, ServiceNow).

Position Telemetry

Location Seattle, WA (Hybrid)
Department Applied AI
Employment type Full-time
Workplace type Hybrid
Salary range
$190k – $270k annual package