Senior Software Engineer

Harvey Nash; Vancouver; Presencial; CAD 140 000 - 170 000; Tempo integral

Resumo da oferta

Harvey Nash in Vancouver seeks a Software Development Engineer, Cloud Platform to build the infrastructure that securely delivers vision software to customer devices. You will create a trusted channel, cloud deployment pipelines, and internal tools for the broader engineering team.

You’ll own architecture decisions, drive security certifications, and implement cost-aware, multi-tenant systems that scale across enterprise customers while balancing on-prem and cloud components.

Qualificações

• 5+ years building backend or platform infrastructure in production.
• Cloud experience (AWS preferred) and infra-as-code.
• Practical networking and secure connectivity with auth controls.
• Experience with multi-tenant IAM and security audits.
• Fluency in Python and data pipelines; observability and cost focus.

Responsabilidades

• Build the connectivity layer to devices with a secure, trusted channel to deployments, telemetry and support.
• Design enterprise-scale account and permissions structures for IT management.
• Migrate deployment systems to the cloud with versioned updates and RBAC.
• Add role-based access control and controlled update workflows for customers.
• Bridge field data and lab data for triage, labeling and model training.
• Operate infrastructure with cost tracking and data lifecycle policies.
• Turn internal tools into customer-facing products with documentation and status tracking.

Conhecimentos

Backend infra

Cloud computing

AWS

IaC

Networking

IAM & auth

Python

Observability

Cost optimization

Ferramentas

Terraform

Kubernetes

Docker

Git

Descrição da oferta de emprego

Rate: $140,000 - $170,000 / Annum

The Role

We’re looking for a Software Development Engineer, Cloud Platform to build the infrastructure that gets vision software onto customer devices: securely, repeatably, and without a member of our team on the phone with someone’s IT department.

Getting software to a device today means threading through customer firewalls, remote-access tools that enterprise security teams increasingly reject, and internal systems only a handful of people know how to drive. You’ll replace that with a real platform: a secure, trusted channel to every deployed device, cloud-hosted deployment infrastructure with proper access control, and internal tooling the rest of engineering can use without a guide.

This is a high-ownership role on a small team. The road-map is written but not finished, and you’ll have significant say in the sequencing, the architecture, and what we choose not to build.

What You’ll Do

• Build the connectivity layer to our devices. Every system in the field needs a secure, trusted channel back to us for deployment, telemetry, and support. You’ll design that channel so it becomes the single point of contact between company and a customer’s cell, shrinking our footprint on their network to something a security team can approve in one conversation, ending our reliance on third‑party remote‑access tools, and handling the reality that some of our best customers run disconnected sites.
• Make Company easy for enterprise IT to approve. You’ll build an account and permissions structure that works at enterprise scale, so IT managers can administer their own users and share resources across teams, and you’ll own the infrastructure work behind the security certifications our customers expect.
• Move deployment to the cloud. You’ll migrate the systems we use to deploy software and manage our device fleet off ageing on‑premise hardware and into the cloud, without a break in service. You’ll own how an update gets built, versioned, distributed, and authorized, working with our on‑device platform engineer on how it gets applied. Along the way you’ll add role‑based access control, so pushing an update to a customer’s production cell is a deliberate act by someone authorized to do it, not an accident waiting to happen.
• Close the loop between the field and the lab. Devices see failure modes our test data never will. You’ll build the path that gets data off a device and into our systems for triage, labelling, and debugging, so field reality feeds back into model training instead of dying in a support ticket.
• Run the infrastructure like a business. Training and storage costs scale with our success, and right now we keep everything forever. You’ll build cost attribution down to the team, customer, and engineer, put data lifecycle policies in place, separate development from production, and evaluate additional providers for both cost and capacity headroom.
• Turn internal tools into products. Our internal tooling works, but mostly for the people who wrote it. You’ll give it a front door, documentation good enough that another team can use it without our help, sensible authentication, and status tracking that alerts on failed jobs instead of letting them sit unnoticed.

What You Bring

• 5+ years building backend or platform infrastructure in production, with meaningful ownership of systems other engineers depend on.
• Strong cloud experience (AWS preferred) and fluency with infrastructure-as-code; you’ve made a migration succeed, ideally one where the legacy system couldn’t be turned off on day one.
• Practical networking and secure connectivity depth (VPN and tunnelling, NAT traversal, certificate and identity management), plus the judgement to design something a customer’s security team will actually approve.
• Real experience with authentication, authorization, and multi‑tenant account models, including access control in environments where a mistake has physical consequences.
• Comfort operating under a compliance framework such as SOC 2, ISO 27001, or TISAX, and a willingness to own the infrastructure side of an audit rather than route around it.
• Python fluency and experience with job orchestration or data pipelines, plus instincts for observability and cost.

Bonus

Managing fleets of edge or embedded devices, air‑gapped or disconnected deployments, containerized delivery to customer hardware, or prior work in robotics, manufacturing, or another industrial setting