Staff Software Engineer, AI Engineering

Our client is a well-funded healthtech company serving over 140,000 independent medical practices across the US — think the obstetrician's clinic, the physiotherapy practice, the small surgical group. They achieved unicorn status in 2022 and closed a $250M funding round in late 2024, with a significant portion of that mandated specifically to accelerate their AI transformation. The engineering culture is strong — average tenure on the team is six to seven years, which says a lot — and this is a rare opportunity to build AI-native infrastructure from the ground up in one of the most complex and high-stakes domains in the US economy.

Our client is building a new class of AI-native systems where intelligence is embedded directly into application workflows — not bolted on as a standalone feature. These systems combine large language models, machine learning models, deterministic logic, event-driven architectures, and stateful workflow execution to automate some of the most complex operational processes in US healthcare.

They're looking for an AI Engineer to design and build production-grade AI systems that can reason, make decisions, and take action within real clinical and billing workflows.

What you'll work on

You'll be operating primarily in the AI orchestration layer, building the intelligence infrastructure that powers their next-generation platform. This includes:

• Designing and implementing multi-step, stateful workflows that coordinate models, tools, and APIs — including long-running execution, failure recovery and retries, partial execution and resumption
• Building LLM-powered applications for reasoning, extraction, and decision support — structured outputs, tool usage, prompt-driven workflows, both real-time and asynchronous
• Designing hybrid AI architectures that combine LLMs (reasoning and language understanding), ML models (prediction and scoring), and deterministic logic (validation and enforcement) — using each appropriately based on the task
• Building event-driven systems with asynchronous processing, idempotent execution, retry and failure handling, and state change-triggered workflows
• Developing retrieval and context-aware systems — vector search, embedding-based approaches, RAG pipelines, internal and external data integration
• Ensuring systems are production-ready: logging, tracing, monitoring, resilience to failures and edge cases

Specific capabilities you'll build

• AI orchestration workflows using LangGraph, LangChain, AutoGen, or custom orchestration frameworks
• Tool-calling and agent architectures integrated with APIs, databases, knowledge bases, and RAG pipelines
• Reasoning workflows covering planning, tool execution, and validation loops
• RAG systems including vector databases, embeddings pipelines, and retrieval workflows
• AI safety and guardrails — prompt safety, hallucination mitigation, deterministic validators
• AI-driven workflows for claim generation, note summarization, decision support, and automated actions

What they're looking for

• 5–10+ years of software engineering experience
• Strong backend development experience — Python preferred
• Experience building distributed and asynchronous systems
• Hands-on experience with RAG pipelines and production LLM systems
• Experience with workflow orchestration concepts and event-driven architectures
• System design experience for reliability and scale
• 
  

What will make you stand out

• Production experience with LangGraph or LangChain orchestration frameworks
• Experience with vector databases (Pinecone, ChromaDB, or similar)
• Experience deploying models on Vertex AI or equivalent ML platforms
• Streaming systems experience — Kafka, Pub/Sub, or similar
• Prior work building AI-first or ML-driven applications

What makes this role different

This is not an AI integration role. Our client is building AI systems as core product infrastructure — systems that execute multi-step workflows driven by AI, maintain state across complex processes, combine multiple forms of intelligence into a single system, and continuously improve through data and feedback.

Candidates whose experience is limited to prompt engineering, chatbot integrations, or traditional REST backend development are unlikely to be the right fit.

Tech stack

Python, OpenAI/Gemini or similar LLMs, LangGraph/LangChain, Vertex AI, Kafka/Pub/Sub, Pinecone/ChromaDB, Docker, Kubernetes, cloud data warehouses, object storage

Location & model

Fully remote. Canadian candidates 

Compensation:

$180,000-$220,000 CAD + Benefits