Sentis
Sentis

Forward Deployed Engineer, Capture Technology and Infrastructure

$170–250kSan FranciscoOn-siteSeniorFull-time

Posted 11 days ago

The role in short

Build and deploy computer vision capture software and edge infrastructure for hospital ICUs, working on-site to integrate with clinical systems and ensure production stability.

About Sentis

Sentis is capturing the most accurate state of the human body. We build at the intersection of physical AI and healthcare. We are live in US ICUs, piloting with academic medical centers, and scaling from single units to full-hospital deployments. Our ambition is simple: shaping the future of medicine.

The Role

Every ICU in the country runs on bedside monitors from different vendors, in different configurations, showing data that is almost never captured. Our technology reads it anyway, from any monitor, in real time, installed in ten minutes.

You are the engineer who builds it. You own the capture software and the infrastructure it runs on, hands on, from the computer vision model on the device to the fleet and the pipeline that carries the data to the cloud. You work with our Platform Architect, who sets the end-to-end architecture and has run device fleets at industrial scale, and directly with our founders, Tim Topper and Christin Seltmann.

The work continues after the code is merged. You go into the ICU, install nodes, watch them fail in ways no test predicted, and fix the cause the same week. What you learn at the bedside goes straight back into the product.

What You Will Do

Build the capture technology. You develop the computer vision pipeline that reads physiological data from bedside monitor displays across thousands of layouts and vendors. You push accuracy, latency and sampling rate further, and you extend what we capture from numeric vitals.

Build and run the infrastructure. You implement device provisioning, secure onboarding, over-the-air updates, remote diagnostics and fleet monitoring for nodes that run unattended inside hospital networks for months. You build the pipeline that moves clinical data from the bedside to the cloud in real time, and you design for unreliable networks, graceful failure and automatic recovery.

Own deployments end to end. You take each hospital from the first network and security review to stable production. You scope the rollout with clinical engineering and IT, install on the unit, validate the captured data against the monitor, and hand over a system that runs without you.

Work inside the hospital's systems. You integrate with hospital networks, EHRs and interoperability standards such as HL7 and FHIR. When a CISO or a clinical engineer has a hard question, you answer it yourself.

Measure, then improve. You define how capture accuracy is measured and build the evaluation harness for it. You use error analysis and monitoring from live nodes to find what fails and raise the numbers.

Turn every deployment into product. What breaks at one site becomes a fix for all sites. You turn field lessons into tooling, tests, install playbooks and reusable components, so the tenth hospital takes a fraction of the effort of the first.

Build security and compliance in. You protect patient data at every layer: encryption in transit and at rest, access control and audit trails. You carry the engineering work behind HIPAA, SOC 2 and hospital security reviews.

Build like it is 2026. You use AI coding tools such as Claude Code as a core part of how you work, and you ship with the speed and quality of a much larger team.

What Makes This Different

Health AI will only be as good as the data it learns from. The systems you build capture data that has never been captured before, at the bedside of the sickest patients in the hospital. Few engineering problems combine computer vision, edge systems, device fleets, real-time data and direct clinical impact in one product.

You will see your work in use. What you ship runs in an ICU, and you will stand next to the monitor and watch it work.

You will build next to people worth learning from. Christin has scaled and sold a company before, holds a Harvard MBA, and brings years at McKinsey and Siemens. Tim started as a paramedic, where he experienced firsthand how missing data can put a patient's life at risk. He went on to AWS, Columbia and Stanford. Sentis is backed by 776, Valor Equity Partners, and Bling Capital, and supported by senior leaders from Merck and Anthropic.

There is no legacy system here. You build it, and your ownership and your equity reflect that.

You Might Be a Fit If

The bar for this role is high on purpose. We are looking for one of the best engineers of their generation, and we would rather wait than lower it.

  • You are technically exceptional and there is proof: a system you built that others said could not be built on that timeline, a product you created from nothing, open-source work, or competition results
  • Experience as FDE at Palantir, OpenAI, or Anthropic is a huge plus
  • You have at least five years of building and shipping production software, and you have personally taken a hard system from first prototype to stable production
  • You are obsessed. A hard problem absorbs you, and you do not let go until it is solved
  • You are hungry and fast. You measure progress in days, and you would rather ship and correct than wait for certainty
  • You write production-grade Python and C++ and are fluent in embedded Linux and containerized edge deployments
  • You have production experience with computer vision or signal processing, ideally on resource-constrained edge hardware
  • You have put devices into the field and kept them alive: provisioning, OTA updates, secure boot and remote management, on networks you do not control
  • You have built real-time data pipelines with technologies such as MQTT, Kafka and time-series databases on a major cloud platform, ideally Google Cloud
  • You have worked forward deployed, or the equivalent: inside a customer's environment, next to their engineers and operators, owning the outcome. Former technical founders and first engineers fit this well
  • You operate autonomously, stay calm when the stakes are high, and make sound technical calls with incomplete information
  • You explain your work clearly to an ICU nurse, a clinical engineer and a hospital CISO alike
  • You are among the top 1% of AI coding users worldwide. Tools like Claude Code are central to how you design, build and ship
  • Bonus: you have already worked in hospital settings, with clinical engineering, medical devices or EHR integration (Epic, Oracle Health, HL7, FHIR), or under HIPAA, SOC 2 or IEC 62304
  • You are ready to spend about a quarter of your time at hospital sites across the US and occasionally abroad, more during rollouts
  • You are based in the US and authorized to work here without visa sponsorship

We hire for what you can build. Many of the best engineers we know think differently, and neurodivergent candidates are welcome here.

The Environment

You will work in person with Tim and Christin in San Francisco, and on site in the hospitals we deploy in. We are a small team and the pace is high. As one of our first engineers you will shape how Sentis builds.

Compensation

$170,000 to $250,000 base plus meaningful early equity.

Next Steps

We move fast and aim to complete the process within a few weeks: an intro call, a technical deep dive on a real Sentis problem, and a conversation with both founders and our Platform Architect.

With your application, send us the hardest thing you have built and what made it hard. A link or five sentences is enough. We read that before your resume.

Prep for this interview

4 questions likely to come up at Sentis, each matched to a line in this posting.

  1. 1Integrations and APIs
    They wrote: “Bonus: you have already worked in hospital settings, with clinical engineering, medical devices or EHR integration (Epic, Oracle Health, HL7, FHIR), or under HIPAA, SOC 2 or IEC 62304”

    Walk me through connecting our product to a customer's internal system that has no documentation.

    How to answer

    Why they ask: This happens weekly in the job. They want a calm, step-by-step method.

    1. Find who owns the system and get read access to a test environment
    2. Watch real traffic or logs to learn the data shapes
    3. Build a thin adapter with logging around every call
    4. Agree on how failures get reported, and who fixes what

    Tip: Talk about finding the owner of the system first. The code is the easy part.

  2. 2Data and SQL
    They wrote: “You have built real-time data pipelines with technologies such as MQTT, Kafka and time-series databases on a major cloud platform, ideally Google Cloud”

    How would you design a daily pipeline that feeds customer data into our product?

    How to answer

    Why they ask: Tests basic data engineering: scheduling, failures, re-runs and monitoring.

    1. Extract with timestamps or change tracking, not full dumps, when possible
    2. Transform in steps you can re-run without creating duplicates
    3. Validate row counts and key fields before loading
    4. Alert a person on failure, with enough detail to fix it

    Tip: Make re-running safe (same input, same result) and say how you'd know it failed.

  3. 3Coding
    They wrote: “You have at least five years of building and shipping production software, and you have personally taken a hard system from first prototype to stable production”

    How do you keep quality high when you ship fast at a customer site?

    How to answer

    Why they ask: Speed is the job, but broken code at a customer is costly. They want your guardrails.

    1. Tests for the parts that touch customer data or money
    2. Code review even when it's fast, ideally by someone on the product team
    3. Feature flags and an easy way to roll back
    4. Clean up the shortcuts in the next release, and write them down

    Tip: Name the few checks you never skip, and what you skip on purpose.

  4. 4Ownership and working with customers
    They wrote: “There is no legacy system here. You build it, and your ownership and your equity reflect that.”

    Tell me about something you shipped that broke in production.

    How to answer

    Why they ask: Everyone has broken production. They want to hear you own it and change something afterwards.

    1. What broke and who it affected
    2. How you found out and what you did in the first hour
    3. The root cause, said plainly
    4. What you changed so it can't happen the same way again

    Tip: Pick a real one with a cost. Spend most of the answer on what you changed.

Not the one? Get the next one by email

New roles land here every day. Pick what you want to hear about.

FDE Info

Want the job, not just the posting? Learn how FDEs work.

FDE Info is our free weekly newsletter for forward deployed engineers: tools, field notes from real deployments, and how FDE teams hire and pay.

Free. One email a week. Unsubscribe in one click.

Similar roles at other companies

Sentis

$170–250k · On-site

Apply