Full Stack Software Engineer, Applied AI San Francisco
Wonderschool, Inc.
Get more replies from employers
Send a job-specific resume in minutes.
Benefits offered by this job
Job summary
Qualifications
- 5+ years of experience as a full stack engineer.
- Hands-on experience with AI coding agents.
- Proficiency with React/TypeScript and one backend language.
- Experience with relational databases like Postgres.
- Clear communication skills, both written and verbal.
Responsibilities
- Review AI-generated code and provide feedback for improvement.
- Architect codebase for AI legibility and maintainability.
- Build and automate product pipeline with AI agents.
- Debug issues across the entire tech stack.
- Monitor and maintain the production environment.
Skills
Tools
Job description
Wonderschool builds software that helps small business owners earn more money. We started with childcare providers because it is a large, underserved market where technology can have an outsized impact on revenue. But the model scales: we are building a platform and a playbook that will expand into other verticals.
We are heavy users of AI. OpenClaw agents run most of our operations: provider outreach, customer communications, enrollment workflows, compliance tasks. Non-engineers at Wonderschool spin up automations, cron jobs, and API workflows directly from Slack, without touching a line of code. We are now taking the next step: making AI agents reliable enough to build the product itself.
Our team is small and moves fast. We value engineers who are obsessed with outcomes, not process.
The North Star for this role: a system where a user signal (a provider revenue dropping, an enrollment bottleneck, a churn spike) automatically triggers a chain of AI agents that writes the product requirements, designs the solution, builds the code, reviews it, and ships it. No human in the loop for the routine stuff. Humans define what matters, train the agents, and review edge cases.
We are not there yet. You are the person who helps us get there, piece by piece.
- Review and iterate on AI-generated code (from tools like Claude Code, OpenClaw, Codex, Cursor) and feed that review back as training signal to make the agents better over time
- Architect the codebase for AI legibility: clean data models, strong documentation, deprecation of legacy patterns that cause agents to fail
- Build and maintain the automated product pipeline: signal detection, agent-generated requirements, AI-driven development, AI code review, commit
- When agents break in production, you own the diagnosis and fix
- Build tools and workflow infrastructure that let operations, sales, and customer success teams operate the platform themselves, without filing engineering tickets
- Translate what internal teams need into automated, reliable systems
- Help non-engineers understand what is possible and then make it happen
- Build features across the frontend and backend: React/TypeScript on the front, Node.js or Elixir on the back, Postgres underneath
- Own the full lifecycle from requirements to production
- Debug issues across the entire stack
- Actively observe the production environment: usage patterns, agent failures, edge cases, provider behavior
- This role requires a high degree of personal accountability for system health. When things are running well, that is because you have built the monitoring and alerting to catch problems early. Getting there takes commitment.
- 5+ years of experience as a full stack engineer
- Hands‑on experience with AI coding agents (Claude Code, OpenClaw, Codex, Cursor, or similar) in a real development workflow
- Strong product instincts: you can take ambiguous user signals and turn them into concrete technical requirements without heavy PM support
- Proficiency with React/TypeScript and at least one backend language (Node.js, Elixir, Ruby on Rails, or similar)
- Experience with Postgres or similar relational databases
- High ownership mentality: you do not wait to be told something is broken
- Clear, direct communication skills, written and verbal
- Experience building or operating agentic workflows in production
- Experience designing feedback loops that improve AI output quality over time
- Background at a startup or high‑growth company
- Experience with CI/CD, DevOps, or cloud infrastructure
- Experience with REST or GraphQL API design
This is not a 9-to-5 role. At this stage, getting AI agents to ship reliable production code requires close observation of what is actually happening in the codebase. When the system is working well, that monitoring is lightweight and mostly automated. Getting it to that point requires genuine commitment: watching the system, catching failures before they compound, and iterating fast.
We are also not looking for engineers who want to write code in isolation. If enabling non‑technical teammates and removing yourself from repetitive engineering work does not sound appealing, this probably is not the right fit.
Base salary: $140,000+, depending on experience, skills, and location. This role is also eligible for equity and other forms of compensation.
Wonderschool offers a competitive benefits package, including:
- Health benefits with up to 100% coverage for employee premiums and up to 80% for dependents
- Wifi and employee wellness stipends
- Flexible PTO, paid holidays, and mental wellness days
- Competitive parental leave (eligible after 6 months of employment)
- A highly collaborative, mission‑driven workplace environment
San Francisco, primarily in-office.
'/%3e%3cpath%20d='M101.744%2012.4726L102.894%2016.1947L99.0021%2013.2201L101.744%2012.4726ZM105.3%208.40937H100.488L99.0026%203.59473L97.5116%208.41047L92.7%208.40385L96.5965%2011.3834L95.1055%2016.1942L99.0021%2013.2196L101.408%2011.3834L105.3%208.40992V8.40937Z'%20fill='white'/%3e%3cdefs%3e%3clinearGradient%20id='paint0_linear_5593_1381'%20x1='90'%20y1='9.44471'%20x2='99'%20y2='9.44471'%20gradientUnits='userSpaceOnUse'%3e%3cstop%20stop-color='%2300AD70'/%3e%3cstop%20offset='0.998594'%20stop-color='%2300B67A'/%3e%3cstop%20offset='1'%20stop-color='%23DCDCE6'/%3e%3c/linearGradient%3e%3c/defs%3e%3c/svg%3e){kind=small}
