Senior Full Stack Engineer — AI Automation Platform

San Francisco, CA or Vancouver, BC · On-site · Full-time

$200K–$300K USD base + competitive equity

The company

This is an enterprise AI automation platform for teams that want to build and deploy AI agents without asking engineering to wire up every workflow.

Users run agents in Slack, Microsoft Teams, or standalone environments, connected across hundreds of integrations to automate real work.

The product has three pillars:

• Agents: conversational AI assistants that use tools to complete open-ended tasks.
• Workflows: visual drag-and-drop automations for repeatable processes.
• Gumstack: enterprise-grade MCP hosting, observability, and AI governance.

Teams at Shopify, Ramp, Gusto, Instacart, Samsara, and Opendoor are already running the platform in production.

The company was founded in 2023, has 31 employees, and has raised $70M from Benchmark, First Round, and Y Combinator.

The product is built for non-technical users, which raises the bar for clarity, error handling, and trust in the experience.

Why now

The platform is already in production at well-known enterprises. The next problem is not proving the idea.

The next problem is making the product easier to adopt, easier to govern, and easier to debug as usage expands across more teams and more automations.

That shifts the engineering bar from shipping isolated features to building a coherent product surface that can handle complexity without becoming opaque.

The role

This is a senior full-stack role for someone who can own a product surface from first design through production support.

You will work across the frontend and backend, ship features end to end, and carry the outcome after launch.

The scope is intentionally broad because the product surface is broad: agent UX, workflow editing, execution state, integration setup, error handling, observability, and enterprise controls all interact.

The engineering team is small: 6 people in San Francisco and 3 people in Vancouver. Scope is large because the team is small.

If you are strongest on one side of the stack, that can still work if you can genuinely operate on the other side. The baseline is full-stack ownership.

The technical problem

AI automation is only useful when it is understandable, recoverable, and safe to operate in production.

The hard part is not building a demo that calls models.

The hard part is building a system where non-technical users can compose agents and workflows, connect them to external tools, inspect what happened, recover from failures, and trust the result inside an enterprise environment.

That means the frontend, orchestration logic, integration layer, and governance surface all have to be designed together.

When agent behavior is probabilistic, the product still has to feel deterministic enough for users to rely on it.

What you'll own

• End-to-end product features: design, build, test, deploy, and own user-facing functionality across the full stack.
• Agent and workflow surfaces: build the interfaces and backend logic that let users create, edit, launch, and monitor automations.
• Execution visibility: make runs, failures, retries, and system state legible to users and to the team.
• Integration-heavy product flows: ship flows that depend on external systems and still behave predictably when those systems fail or change.
• Enterprise controls: help shape the product experience around observability, governance, and operational clarity.
• Production ownership: debug issues after launch, tighten edge cases, and improve the system based on real usage.
• AI-native execution: use AI tooling to move quickly, but keep the standard high for correctness, maintainability, and user experience.

Who this is for

You are likely a strong fit if you have:

• 3+ years shipping full-stack products end to end, with enough depth to own architecture, implementation, and production behavior.
• Already owned meaningful product scope, not just contributed tickets inside someone else’s system.
• Strong product judgment and the ability to turn ambiguous requirements into shippable interfaces and services.
• Experience building interactive product surfaces and the backend systems behind them.
• Comfort working in systems where the output is partly probabilistic and the product still needs deterministic behavior around it.
• Good instincts around state management, failure states, data flow, and user-visible reliability.
• The habit of using AI tools to accelerate execution without lowering the quality bar.
• The ability to move between frontend details and backend constraints without handoffs becoming a bottleneck.
• Comfort working on-site with a small team and contributing directly to product decisions.

This role is not for you if

• You want backend-only or frontend-only scope.
• You need highly specified tickets before you can start.
• You prefer to avoid production ownership after launch.
• You do not want to work on systems where UI, orchestration, and reliability need to be designed together.
• You are not comfortable operating on-site with a small team.

Compensation and logistics

• Base salary: $200K–$300K USD
• Equity: competitive
• Location: San Francisco, CA or Vancouver, BC
• Work model: 5 days per week in-office; no remote flexibility
• Visa support: H1B transfers and existing US work authorization are accepted; new sponsorship applications from outside the US are not supported
• Employment: full-time

About Aurora

Aurora helps exceptional engineers find the right role at some of the most ambitious startups worldwide.

We work with teams that value high ownership, strong technical standards, and clear scope.