Staff Platform Engineer

Why Charlie Health?

Millions of people across the country are navigating mental health conditions, substance use disorders, and eating disorders, but too often, they’re met with barriers to care. From limited local options and long wait times to treatment that lacks personalization, behavioral healthcare can leave people feeling unseen and unsupported.

Charlie Health exists to change that. Our mission is to connect the world to life-saving behavioral health treatment. We deliver personalized, virtual care rooted in connection—between clients and clinicians, care teams, loved ones, and the communities that support them. By focusing on people with complex needs, we’re expanding access to meaningful care and driving better outcomes from the comfort of home.

As a rapidly growing organization, we're reaching more communities every day and building a team that’s redefining what behavioral health treatment can look like. If you're ready to use your skills to drive lasting change and help more people access the care they deserve, we’d love to meet you.

About the Role

Core Platform is a platform team. Our customers are Charlie Health's other engineering teams, and our job is to make them faster, safer, and more consistent. We don't ship product features. We build the patterns, services, and guardrails that other engineering teams depend on.

As a Staff Platform Engineer on Core Platform, you will own one of the primitives at the heart of Charlie Health's replatforming: the federated GraphQL front door, the event substrate that carries every signal, the identity layer for humans and AI agents, or the real-time world model state layer that gives our services and agents a live picture of care. You will take your area from design through production operation, and the patterns you set become the team's paved roads.

Every engineer is a tech lead, and as a Staff engineer you are a domain owner: people come to you with the hardest technical decisions in your area, and you make calls that hold up across team boundaries. You will work alongside a team of experienced engineers who own the platform primitives. We build with AI as the default. Specs are the primary artifact, agents draft most of the implementation, and spec quality is the clearest craft signal at this level. The quality gates this team builds are what make agent-written code safe in a HIPAA-regulated clinical business.

Responsibilities

Own one or more platform primitives end to end, from architecture and build through production operation and long-term evolution

Build the deterministic quality gates that let agent-written code ship safely: automated end-to-end testing, golden test datasets, spec templates, and schema change safety

Deliver core services on the team's own paved roads, finding the potholes before any product squad does

Write specs that name invariants, error cases, security implications, and acceptance criteria, and orchestrate agent swarms on hard multi-system work

Coach Senior engineers and product squad engineers on spec writing, agent orchestration, and platform patterns

Define the reliability conventions for your area: SLO definitions, observability standards, and incident response practice, including a seat in the on-call rotation

Steward the security and compliance posture of your primitive, and stay aware of what it costs to run

Partner with product squads as they federate into the supergraph and adopt the event substrate during the migration onto the new platform

Design cross-service patterns that product squads adopt, and treat adoption as the measure of success rather than delivery alone

Document the standards in your domain as architecture decision records and bring them through our Architecture Advice Guild

Requirements

7+ years of software engineering experience, including several years designing and operating distributed systems or platform components that other teams build on

Depth in at least one of these areas, enough to own it end to end: GraphQL federation, real-time stream processing (Flink, Spark, KsqlDB, or similar), graph databases (Memgraph, Neo4j, Amazon Neptune, Dgraph, or similar), or authorization systems with ReBAC and RBAC models (Zanzibar, OpenFGA, Cedar, or similar)

Experience with skills, MCPs, harnesses, agentic tooling ecosystems, or agent orchestration for software development

You build with AI agents as part of your daily work. You write specs agents can implement, run agents for real delivery, and know where agent output needs human judgment