Member of Technical Staff (Software Engineer, Connector Platform)
Remote • San Francisco • FullTime
Posted 3mo ago
Remote Work Policy
Fully remote
Employment Type
FullTime
Categories
AI Infrastructure Engineer
About the job
The Connector Platform team is responsible for building the data layer that enables Perplexity's agents to interact with the world's software. This team manages systems that transform hundreds of diverse integrations into a unified, reliable, and well-typed interface for agents. This platform serves as the core knowledge layer, enabling agents to discover, understand, and utilize tools effectively, grounding their reasoning in real, permissioned, and up-to-date enterprise data. By maintaining a knowledge layer above connectors, the platform ensures Computer becomes the single source of truth for institutional knowledge, providing grounded, actionable, and permissioned access to customer systems, which is a key differentiator in the market.
Responsibilities
- Own the design and implementation of the connector runtime, including registration, hosting, and execution of various connector types.
- Build and extend the semantic layer, encompassing tool and entity schemas, capability metadata, and relationship modeling.
- Design tool-discovery and tool-selection surfaces to optimize agent accuracy and context efficiency.
- Enhance agent loop robustness with structured results, partial-failure handling, idempotency, pagination, rate-limit management, and observability.
- Define authentication, authorization, and credential-isolation patterns for connectors, collaborating with Security and Backend Platform teams.
- Develop the connector onboarding path, including schemas, fixtures, and evaluation suites, to ensure measurable quality.
- Establish technical standards for connector reliability and operability, including SLAs, observability, and incident response.
- Partner with product and AI teams to define clear connector interfaces and integration patterns for new agent capabilities.
Requirements
- Experience designing and building backend systems in production (typically 4+ years for mid-level).
- Strong system design skills with a track record of building efficient, reliable, and scalable architectures, ideally involving API integration or platform systems.
- Proficiency in at least one backend language like Python, Go, or Rust, and ability to work in a multi-language environment.
- Hands-on experience with modern infrastructure such as AWS and Kubernetes.
- Depth in one or more of: OAuth and authorization protocols, API/connector or MCP-server development, schema and semantic modeling, or building tooling for LLM-based agents.
- Comfort working in security-sensitive areas and making pragmatic trade-offs.
- Collaborative mindset and eagerness to solve complex, ambiguous problems.