
- B2C
- Scale StageRapidly increasing operations
Forward Deployed Engineer
- $120k – $150k
- |
- |Full Time
In office - WFH flexibility
Not Available
About the job
Forward Deployed Engineer
Location: Los Angeles (hybrid) or remote, US
Reports to: CTO
Type: Full-time, senior individual contributor
Compensation: $120-150k
The role in one paragraph
We are looking for one person who can do the work that currently takes a product manager, a backend engineer, a frontend engineer, an analyst, and six weeks of calendar time — and do it in an afternoon, using Claude. You will sit outside the normal delivery process and report directly to the Executive Team. You will be handed problems, not tickets, and you will be measured on working software in production, not on story points or status updates. If the honest description of your last two years is "I stopped waiting for a team and started shipping," read on.
What you'll actually do
- Take a business problem stated in one sentence by an operator — a pharmacy lead, a clinical lead, a growth lead — and return a working tool, usually the same week. Probably multiple.
- Build and own internal systems end to end: operational dashboards, partner-facing reporting, workflow automation, data pipelines, admin tooling, integrations between systems that don't talk to each other today.
- Integrate across our stack — EHR and clinical workflow, pharmacy fulfillment, e-prescribing, payments, CMS, warehouse and analytics — including vendors whose APIs are undocumented, hostile, or nonexistent.
- Write the spec, build the thing, deploy it, instrument it, and then sit with the person using it and fix what's wrong. No handoffs. No throwing it over a wall.
- Build the internal AI infrastructure the rest of the company runs on: agents, skills, evals, and the guardrails that keep them safe in a HIPAA environment.
- Kill work, not just create it. A meaningful part of this job is replacing a manual process, a vendor invoice, or a recurring report with something that runs itself.
What "fast" means here
This is the part of the job description you should take literally.
- A read-only internal dashboard: same day.
- A net-new internal tool with auth, a database, and a real workflow: 2–5 days.
- A third-party integration: days, not a quarter.
- If your instinct on hearing a request is to schedule a scoping call for next week, this role will make you miserable.
You will not be asked to guess. You'll be asked to build the smallest honest version, put it in front of a real user within 48 hours, and iterate in public.
You're probably a fit if
- You have shipped production software that real people depended on, and you can point at it.
- You are genuinely fluent with Claude or comparable frontier models as a build tool — not autocomplete. You have opinions about context management, agent design, evals, and where these tools break. You can explain a build where the model did 80% of the work and you did the 20% that mattered.
- You have product judgment. Given a vague ask and access to the operator, you can figure out what should actually get built, including deciding not to build it.
- Full stack is table stakes: you can stand up a database, an API, a frontend, auth, and a deploy without asking anyone for help.
- SQL and data modeling are second nature. You reach for the warehouse before you reach for an opinion.
- You are comfortable being the only person on a project, and comfortable showing unfinished work.
- You can work inside a regulated environment — PHI, HIPAA, pharmacy and telehealth constraints — without either ignoring the rules or hiding behind them.
You're probably not a fit if
- You need a written spec, a designer, and a backend partner before you can start.
- Your throughput depends on someone else's sprint ceremony.
- You want to manage people. This role has no reports and isn't a path to one in year one.
- You think of AI coding tools as a productivity add-on rather than a change in what one person can build.
- You need certainty about the roadmap. The roadmap is "whatever is most broken right now," reprioritized weekly.
First 30 days
- Days 1–3: Ship one small, visible internal tool that removes a recurring manual task.
- Days 3–15: Deliver multiple systems that were previously quoted in months. Stand up the AI tooling and eval setup the rest of the company will use. Start making changes in our EMR.
- Days 15–30: Own a series of tickets and succeed more than anyone else on the team
How to apply
Send two things to:
- One page on something you built where the tool did most of the work and you made it good. Include what broke.
- A link to something we can open. A small working tool, not a repo of scaffolding. If you built it with Claude, include the prompts or transcript — we care more about how you steer the model than about your commit history.
No cover letter. We'll respond to everyone who sends a working link.
About the company

Lemonaid Health
- B2C
- Scale StageRapidly increasing operations
Employees joined from
Similar Jobs




