The company's app, tracking devices and pet GPS tracker help members look after family, pets and belongings through location sharing, safe driver reports, and crash detection with emergency dispatch.
As of September 30, 2026, the company served about 107.7 million monthly active users in more than 180 countries. It has over 500 employees and works remote-first.
The AI Platforms team builds the shared base that lets the company adopt AI quickly and safely. Its scope covers routing to models, the MCP gateway and connectors, agent runtime and orchestration, memory and knowledge stores, evals and observability, and cost controls.
The team works with AI Dev Tooling and Agentic PDLC, which are changing how the company builds with agents, and is extending the same base to AI features for members.
This Staff role reports to the Director of AI Platforms. You will carry internal platform work into member-facing features, supporting every engineer and agent at the company.
The company is building an AI-native organization. AI use in interviews depends on the role, and your recruiter will explain what is expected at each stage. Undisclosed AI use that was not discussed with or approved by your recruiter may affect your candidacy.
Responsibilities
Set the overall direction and build paved paths that help teams adopt AI-native engineering practices
Build the agent runtime and orchestration platform, including identity, authorization, and audit for scheduled and unattended agents
Run the MCP gateway and connector infrastructure so agents reach internal data, APIs, repositories, and SaaS tools only through managed, auditable connections
Maintain the LLM gateway, covering model access, routing, authentication, rate limits, token budgets, and cost attribution by team
Build eval and regression tooling, including golden datasets, offline evals, CI hooks, and result storage, and treat quality regressions as top-priority alerts
Set up observability for prompt and response tracing, cost per request, quality signals, and drift detection
Work with Security, Privacy, and Legal to build data handling, tenant isolation, and auditability into the platform from the start
Define how product teams access, run, and monitor production AI inference at member scale, including serving, latency, reliability, and provider management
Own your production components, including SLOs, dashboards, deployment pipelines, and incident response
Build memory, retrieval, and knowledge infrastructure that agents can share
Make the preferred approach the easiest one with SDKs, templates, reusable skills, documentation, and internal champions
Work with AI DevX and Agentic PDLC to turn their needs into shared primitives rather than one-off builds
Mentor engineers on agent architecture and evaluation, and write design docs and standards that other teams can build on
Research longer-term memory, retrieval, and shared knowledge capabilities as the platform's needs grow
Requirements
8+ years of engineering experience, including 3+ years building platforms or infrastructure that other engineers use
Experience building and running production AI systems, such as agentic workflows, LLM integrations, retrieval, and evaluation
Working knowledge of MCP and agent orchestration, including how agents connect to internal systems under proper controls
Strong backend and cloud skills: distributed systems, APIs, infrastructure as code, and reliability and observability practices
Experience running AI inference in production, weighing latency, reliability, provider management, and cost
Good judgment about which work deserves a shared platform investment and which does not
Clear written communication, with design docs and standards detailed enough for other engineers and their agents to build against
Bonus: experience with non-human identity, service-account, or agent authorization models
Bonus: experience building or operating an agent orchestration or deployment platform, or an MCP-style connector gateway
Bonus: familiarity with SOC 2, GDPR, or similar compliance frameworks
Bonus: background in ML infrastructure or model observability
Bonus: experience with consumer-scale products or privacy-sensitive data
You use AI coding tools (the company supports Claude Code, Cursor, and GitHub Copilot) daily for substantive work such as analysis, coding, refactoring, testing, navigating codebases, and documentation, not only research or autocomplete
You take ownership of AI-generated code and review it the same way you would any other pull request
You use AI well enough to take on work that would otherwise need a larger team, and you share prompting approaches and help teammates adopt AI tools
You keep up with fast-changing tooling and recommend changes when they would meaningfully improve how the team works
Conditions
The US base salary range for this role is $149,000 to $271,500 USD. Final pay depends on background, job knowledge, skills, and experience, and may vary by one of three US geographic tiers
This is a remote-first role. It can be performed remotely from anywhere within the US, regardless of any location listed above
The company is an equal opportunity employer and does not discriminate on any legally protected basis
Benefits
Total compensation includes Restricted Stock Units (RSUs) as well as a full benefits package
Medical, dental, vision, life and disability insurance, with 100% of the employee premium paid
Paid parental leave
401(k) plan with company matching
Mental wellness program and employee assistance program
Flexible PTO, company holidays throughout the year, and summer and winter shutdowns
Learning and development programs
Equipment, tools, and reimbursement for a productive home office
Free membership in the company's app and free tracking devices