Forward-Deployed Engineer

PathwaysX, Inc.; Boston (MA); Hybrid; USD 180.000 - 240.000; Vollzeit

Benefits dieser Stelle

Health Insurance

Equity

Flexible Work Hybrid

Paid Time Off

Learning & Budget

401(k) with match

Zusammenfassung

PathwaysX is seeking a Senior ML Engineer to embed with Halcyra’s health-system customers, shaping LLM-powered clinical and administrative workflows from concept to production.

You will work with clinical operations teams, integrate with EHR/security environments, and own deployments through go-live, deciding which parts of a workflow the model touches.

Qualifikationen

• Must have built and operated an LLM or retrieval service in production.
• Can explain a concrete latency, accuracy or cost trade-off they made in that system.

Aufgaben

• Lead on-site technical discovery with hospital operations, IT and security teams.
• Build and deploy LLM and retrieval services integrated with Epic/Cerner via FHIR and HL7.
• Design evaluation sets with clinical reviewers and hold the go-live bar on them.
• Own customer security reviews, PHI handling and deployment runbooks.
• Feed field learnings back into the core product roadmap.

Kenntnisse

LLM deployment

Production systems

Security reviews

Customer discovery

Tools

FHIR

HL7

Epic

Cerner

Jobbeschreibung

Embed with Halcyra's health-system customers to take LLM-powered clinical and administrative workflows from a scoped idea to production. You will sit with clinical operations teams, integrate with their EHR and security environment, and own the deployment through go-live, deciding with the customer which parts of a workflow a model should and should not touch.

Health Insurance

Medical, dental and vision; 100% employee premium covered.

Equity

Meaningful stock options, 4-year vest with 1-year cliff.

Flexible Work

Hybrid: 3 days a week in the Boston office or on-site with customers.

Paid Time Off

Learning & Budget

$3,000 annual learning and conference budget.

Retirement Plan

401(k) with 4% company match.

02 Responsibilities

• 01 Lead on-site technical discovery with hospital operations, IT and security teams.
• 02 Build and deploy LLM and retrieval services integrated with Epic/Cerner via FHIR and HL7.
• 03 Design evaluation sets with clinical reviewers and hold the go-live bar on them.
• 04 Own customer security reviews, PHI handling and deployment runbooks.
• 05 Feed field learnings back into the core product roadmap.

What we evaluate against.

Applied Technical Depth

Must have

• Has built and operated an LLM or retrieval service in production.
• Can explain a concrete latency, accuracy or cost trade-off they made in that system.

Nice to have

• Writes infrastructure as code as well as application code.

Must have

• Has broken an ambiguous customer request into phased deliverables with clear acceptance criteria.

Nice to have

• Explicitly deferred or cut part of a request, with reasons.

Customer & Domain Discovery

Must have

• Has run on-site discovery with operational users, not just IT.
• Can name a workflow constraint discovered in the field that changed the build.

Nice to have

• Healthcare or other regulated-domain exposure.

Product & Solution Judgment

Must have

• Has set a scope boundary for a model-backed feature based on evaluation evidence.

Nice to have

• Has fed field learnings into a core product roadmap.

Execution Under Ambiguity

Must have

• Has shipped to a hard customer deadline under changing requirements.

Nice to have

• Has led a customer security or compliance review.

04 Team dynamics

Team Structure

Customer-embedded pods: 2 FDEs, a solutions architect and a clinical informaticist per health-system account.

Operating Rhythm

Monday account reviews, async daily updates, fortnightly deployment retros.

Decision Making

Pods decide scope with the customer; architecture changes go through a weekly design review.

Ambiguity & Pace

Each new health system brings different EHR integrations and security reviews.

Ownership

Pods own a customer deployment end to end, from discovery to go-live.

Customer Obsession

Engineers spend real time inside hospital workflows, not just in meetings.

Transparency

Evaluation results, including failures, are shared with customers.

Learning Culture

Monthly clinical-AI reading group and a generous learning budget.

First 30 days

Ship a first fix into a live customer deployment and complete a shadow week inside a customer's utilisation-review team.

90 days

Take that deployment to go-live against an agreed accuracy and turnaround bar.

Outcomes

Become the pod lead others copy for customer discovery and deployment quality.