Senior Elixir Engineer

ABOUT US
Doctors spend too much of every consult typing. Medi Apps started with AI clinical transcription: we capture the consultation and turn it into a structured clinical note, with diagnosis codes and forms filled in, ready for the doctor to review and sign off.

We are live in clinics in Indonesia today, work with clinicians every week, and ship based on what they tell us.

Now we're expanding beyond the consultation room into the wider hospital: patient registration, informed consent, surgery and operating notes, prescriptions and pharmacy, and the tasks and handovers that connect clinical teams. The goal is one platform that runs a clinical team's day, with AI handling the paperwork and clinicians making the decisions.

The platform is built on Elixir and Ash end to end. You'll work remotely from Australia with a team based in Jakarta (UTC+7), so our working days overlap comfortably.

WHAT YOU'LL WORK ON

• Consultation capture: recording that works live during the consult and keeps working when the connection doesn't.
• Clinical and hospital workflows in Phoenix LiveView: patient charts, notes, surgery, prescriptions, consent, tasks.
• Domain modelling, data security and access control for each clinic.
• Background processing and AI features where the clinician always stays in control.
• Integrations with clinic and hospital systems.
• AI features in the product: turning conversations into notes, codes and forms, with tracing and evaluation, and the clinician always in control.
• The data loop behind our models: annotation tools for clinical reviewers, training data pipelines, quality measurement, and rolling improved models into production.
• Raising the bar: reviews, specs, tests, production support, and mentoring as the team grows.

OUR STACK
Elixir · Phoenix LiveView · Ash Framework · Oban · PostgreSQL · AWS · Grafana stack · Langfuse · Auth0 · Python · ML

HOW WE WORK

• AI-assisted, human-accountable. We use AI coding and review tools heavily; an engineer owns every merge and every deploy.
• Spec first for larger features, small PRs, tests with every change.
• Security and patient privacy are part of everyday engineering, not an afterthought.
• Direct line to users. You will see your work used by doctors and hear their feedback.

YOU HAVE

• 4+ years building production systems, with 2+ years of production Elixir.
• Strong judgement with AI coding tools: you use them daily to move faster, review their output critically, and take full ownership of what you ship.
• Solid OTP fundamentals: processes, supervision, GenServers, failure handling.
• Production Phoenix LiveView experience.
• Strong PostgreSQL and data-modelling skills (Ecto or Ash).
• Testing habits: you write ExUnit tests without being asked.
• Clear written English; comfortable working async with a team 3–4 hours behind.
• Based in Australia with the right to work here.

NICE TO HAVE

• Ash Framework (we will teach it if you're new to it).
• Audio or media processing, or offline-first web apps.
• Healthcare software, HL7/FHIR, or HIS integration.
• AWS, observability (Grafana stack), LLM tracing (Langfuse)
• Multi-tenant SaaS, RBAC, security or compliance work.
• ML in production: speech recognition, model fine-tuning, evaluation, or data annotation pipelines. You don't need to be an ML engineer, but curiosity about how the models work helps.
• Understanding Bahasa Indonesia: many of our users and part of our team speak it.

WHAT WE OFFER

• Competitive package based on experience, inclusive of superannuation
• Fully remote within Australia
• Early-stage team: real ownership, and your decisions shape the product and how we build it.
• Challenging, fun problems at the edge of AI and healthcare. We actively explore the latest AI techniques and tools, and ship the ones that work.
• Work that matters: less admin for doctors, more time for patients.

HIRING PROCESS

1. Intro call (30 min)
2. Technical conversation with an engineer: walk through something you've built (60 min)
3. Pairing session on a realistic task (≈1–2 hrs)
4. CTO call (30 min)

TO APPLY
Send your CV or GitHub, plus a few lines on an Elixir system you're proud of, to randolph@medi-apps.ai