Overview

We are looking for an Intermediate AI Engineer with robust Python development, LLM deployment, and DevOps skills to help build, deploy, and maintain production AI systems for medical imaging operations.The role involves managing locally hosted large language models (LLMs) for medical report generation, developing real-time AI voice agents for patient scheduling, and building scalable, secure AI infrastructure.Our AI systems will support multiple medical imaging locations, with projected scheduling volumes of up to 5,000 patients per day. Key Responsibilities1. Local LLM Deployment & Management

- Deploy, configure, and maintain locally hosted LLMs for medical report generation.
- Optimize model inference performance, GPU utilization, latency, and throughput.
- Work with inference frameworks such as vLLM, Ollama, or NVIDIA Triton.
- Manage model upgrades, versioning, benchmarking, and deployment.
- Monitor model performance, reliability, and infrastructure utilization.
- Implement secure model serving and access controls.
- Support evaluation and validation workflows for AI-generated medical reports.

2. AI Voice Agents & Automation

- Develop and maintain real-time AI voice agents using Python.
- Integrate speech-to-text, LLM, and text-to-speech technologies.
- Implement real-time audio streaming, conversation management, and interruption handling.
- Build scalable infrastructure supporting 100+ concurrent AI calls.
- Integrate AI agents with Davis RIS APIs for patient scheduling.
- Support automated appointment booking, cancellations, and gap-filling workflows.

3. Backend Development & API Integration

- Develop backend services using Python, FastAPI, and asynchronous programming.
- Build secure REST APIs and integrations with healthcare systems.
- Implement background workers, task queues, and scheduling services.
- Work with PostgreSQL, Redis, and distributed systems.
- Ensure reliable API communication, error handling, and concurrency management.

4.



DevOps & AI Infrastructure

- Deploy and maintain AI applications using Docker and cloud or on-premises infrastructure.
- Build and manage CI/CD pipelines.
- Configure GPU-enabled environments for local LLM inference.
- Implement infrastructure monitoring, logging, and alerting.
- Manage application scaling, uptime, backups, and disaster recovery.
- Troubleshoot infrastructure and production AI issues.
- Optimize infrastructure performance and operational costs.

5. Healthcare Data Security & Reliability

- Implement appropriate security controls for sensitive patient information.
- Support PHIPA-aligned privacy and security requirements.
- Maintain audit logs, access controls, and secure API integrations.
- Ensure AI-generated medical reports undergo appropriate clinical validation and approval before release.
- Collaborate with AI developers, radiologists, and IT teams to improve system reliability.

Required Technical Skills

- Strong Python programming experience.
- Hands-on experience deploying and serving LLMs.
- Experience with GPU infrastructure, CUDA, and model inference optimization.
- Familiarity with PyTorch, Hugging Face Transformers, vLLM, or similar technologies.
- Experience with AI agents, LLM APIs, and orchestration.
- FastAPI, REST APIs, and asynchronous programming.
- Docker and Linux server administration.
- Experience with cloud or on-premises production deployments.
- CI/CD pipelines and Git.
- Understanding of networking, concurrency, and distributed systems.
- Experience with production monitoring and troubleshooting.

Nice to Have

- NVIDIA GPU infrastructure and NVIDIA Triton Inference Server.
- Quantization, batching, and LLM performance optimization.
- AI voice technologies, WebSockets, SIP, and telephony APIs.
- Kubernetes and Terraform.
- PostgreSQL, Redis, Celery, or Temporal.
- Healthcare IT systems, RIS/PACS, HL7, and FHIR.
- Experience with medical AI applications and PHIPA requirements.

Expected hours: 40.0 per week

Benefits:

- On-site parking

Work Location: In person

📌 AI Developer (Mississauga)
🏢 MERIVALE MEDICAL IMAGING
📍 Mississauga