Software Engineer

As a Senior Engineer at Magical, you’ll build and own the systems powering our AI agents. You’ll set technical direction for projects, drive delivery, enhance your team’s effectiveness, and take end-to-end ownership of core systems.

Our platform orchestrates long-running, high-stakes healthcare workflows-managing queues, run state, browser automation, model calls, evaluations, observability, and human-in-the-loop recovery.

You’ll help transform unstructured customer inputs-SOPs, decision trees, portal behaviors, and edge cases-into dependable automations. You’ll build systems that can draft, test, and enhance workflows, catch failures, recommend edits, and know when human intervention is needed. You’ll also contribute to improving runtime reliability, workflow pathways, evaluations, and specialized agents for critical healthcare tasks.

This is a hybrid role based in our San Francisco office.

What You’ll Do

• Design and build backend and full-stack systems for AI agents and healthcare automations
• Set technical direction within a project or product area, especially when requirements are ambiguous
• Build features for automation creation, workflow pathways, testing, editing, and production run reporting
• Improve runtime reliability through systems for run state management, queues, idempotency, error classification, alerts, and dashboards
• Own critical services from architecture through deployment, monitoring, and iteration

• You have 5+ years of software development experience.
• You’ve built and owned backend systems deployed in production
• You can turn ambiguous product requirements into a clear technical plan
• You have strong systems design skills-distributed systems, databases, APIs, queues, async workflows, and observability
• You’ve integrated AI tools into your development workflow and are willing to experiment with new approaches
• You have a strong bias for action-you ship, learn, prioritize ruthlessly, and iterate quickly
• You take full ownership of your systems and enjoy mentoring teammates

Even Better If

• You have experience building platform, infrastructure, or workflow systems at scale
• You have built agentic products or agent-based workflows
• You’ve integrated LLMs or AI models into production backend systems
• You have a background in event-driven architectures, workflow engines, observability, or runtime reliability
• You have experience in healthcare operations or revenue cycle management

We’re building the best self-serve agentic automation platform for healthcare, and we’re just getting started. Come join us!

Skills: Application Programming Interface (API), Artificial Intelligence (AI), Artificial Intelligence (AI) Agents, Automation, Distributed Computing, Distributed Databases, Healthcare, LinkedIn, Medical Coding, Patient Care, Patient Care Authorizations, Production Systems, Reliability Engineering, Reporting Dashboards, Revenue Management, Sales, Software Engineering, Standard Operating Procedures (SOP), Systems Reliability, Test Plan/Schedule, Use Cases, Web Browsers

Magical is an agentic automation platform bringing state-of-the-art AI to healthcare-delivering AI agents that actually work in production.

We believe the future of healthcare isn’t about replacing people, but about multiplying the best of them. Magical empowers healthcare organizations to scale their top talent across every workflow, enabling operations to run with greater accuracy, efficiency, and care.

We’re building “AI employees” that automate repetitive, time-consuming workflows while keeping people in control. Our focus is the $4 trillion healthcare industry, burdened with administrative complexity. By automating critical processes like prior authorizations and medical coding, we help providers dedicate more time to patient care.

Our Traction

• Customers are expanding from initial workflows into broader use cases as they see reliable results in production.
• 7-day proof-of-concepts demonstrate real value quickly, in an industry where months is the norm.
• Our self-healing automations deliver production-grade reliability at scale.

We’re backed by Greylock, Coatue, and Lightspeed, having raised $41M. Our founder, Harpaul Sambhi, is a second-time founder who previously sold his first company to LinkedIn.