Full-Stack Machine Learning Engineer

Signal 1•Toronto, ON

37d•Hybrid

About The Position

Signal 1 is building an AI Management System to help health systems operate scalable and valuable AI programs. The Agent Control Plane is a new extension of their flagship product, designed to manage, evaluate, continuously improve, govern, and interact with AI agents within a hospital setting. This role involves a zero-to-one build, co-developing with leading health systems, and offers significant ownership. The ideal candidate has experience taking ML-powered products from inception to production users and wants to take on more responsibility. The team consists of ambitious and passionate engineers, machine learning scientists, and product people who work hard and move fast to improve healthcare AI. The office is in Toronto, with 1-2 days of in-office collaboration per week.

Requirements

3+ years of industry experience shipping software, with a substantial portion on applied machine learning.
Track record of building production AI systems, including LLM-powered or ML-powered features, operating them at scale, and handling issues when they misbehaved.
Depth in AI and backend engineering: strong Python for services and ML pipelines, working fluency with cloud infrastructure and CI/CD.
Working understanding of frontend development (TypeScript, React, or equivalent) with enough hands-on experience to work across the stack.
Deep hands-on data experience: built data pipelines, cleaned messy datasets, designed schemas, and understand data quality impacts.
Zero-to-one shipping experience: taken at least one product or major product area from an empty repo to real users.
History of novel solutions: ability to describe problems with no established approach and the solutions created, why they worked, and how they were validated.
Strong product mindset: user-centric approach driving technical decisions.
Comfort navigating uncharted waters: making good decisions with incomplete information, timeboxing bets, and course-correcting quickly.
High agency and resourcefulness: finding information, unblocking oneself, and seeking help when needed.
Clear, succinct communication: keeping stakeholders informed and translating between technical and clinical audiences.
AI-native way of working: daily use of AI tools for coding, research, and prototyping.
High bar for craft: focus on both functionality and user experience.

Nice To Haves

Experience with healthcare data (FHIR, HL7, EHR integrations) or other regulated, privacy-sensitive domains.
Ownership of a product shipped to production, with responsibility for iteration towards product-market fit.
Experience building agentic systems that operate on large-scale data in production.
Experience conducting AI research, with peer-reviewed publications.

Responsibilities

Own features end to end: Take a problem from ambiguous idea to production, including designing the data model, building pipelines and backend services, and working across the frontend.
Build the AI core of the product: Design and implement systems for grounding agents in hospital context, evaluating their behavior, and converting agent activity into actionable insights for continuous improvement.
Work with messy, real-world data: Ingest and normalize agent telemetry, healthcare data standards like FHIR, and clinician feedback into reliable datasets.
Prototype with design partners: Build demos and pilots with health systems, observe usage, and integrate learnings back into the product.
Invent where there is no playbook: Create solutions for new challenges in agent observability and evaluation in healthcare.
Raise the bar: Keep patterns for testing, evaluation, logging, and reliability ahead of the state of the art and set standards for the team.