Full Stack Software Engineer (AI-Native & Agentic Workflows)

Full-time Permanent Mid-level Senior

Posted 02 Oct 2026

Description

About the Company

Founded in Hamilton in 2019, Space Ops Inc. is a technology development company on a mission to make money movement and online commerce faster, smarter, and more secure for everyone in Canada.

We started Space Ops with a simple belief: the software that runs finance and commerce should be as modern, fast, and intelligent as the people who depend on it. In just seven years, we've grown into a trusted engineering partner for 100+ Canadian businesses.

Today, the platforms and apps we build quietly serve more than 20 million Canadians — processing payments, powering storefronts, and protecting transactions around the clock. We build on the latest and greatest technology, and we've successfully woven AI and automation into everything we do.

About the Role

Building software has fundamentally changed. We are not looking for engineers who occasionally treat LLMs like an auto-complete widget or a glorified Stack Overflow. We are looking for high-leverage builders who have completely rebuilt their engineering workflows around frontier AI systems, agentic loops, and automated verification harnesses.

In this role, you will architect critical systems, ship high-velocity production code, and set the technical standard for what it means to build software with an AI-native operational cadence. You will design pipelines where multi-model orchestration, deterministic test suites, and autonomous coding agents work together to compress delivery timelines from quarters to days.

What You’ll Be Doing





- Lead System Architecture & Execution: Own end-to-end design and implementation for complex, scalable services—maintaining uncompromising standards for security, performance, and code quality.
- Architect AI-Driven Pipelines: Design and operate custom multi-model orchestration pipelines. Route reasoning tasks, structural migrations, code refactors, and test generation dynamically across different frontier models based on latency, context window, and capability thresholds.
- Harness Agentic Systems: Deploy and supervise multi-step agentic coding workflows, Model Context Protocol (MCP)servers, and autonomous background execution loops to automate repetitive architectural patterns and migration workloads.
- Elevate Automated Testing & Verification: Build ironclad automated test harnesses (unit, integration, end-to-end, and property-based) that serve as absolute ground truth and guardrails for code produced across human-AI pair programming workflows.
- Scale Engineering Culture: Mentor team members on tooling, context management, prompt engineering for code, and repo structuring optimized for autonomous tools.

What We’re Looking For

- Deep Production Experience: 8+ years of professional software engineering experience designing, shipping,



and maintaining mission-critical backend or full-stack distributed systems in production.
- Frontier Model Mastery: You possess a granular, intuitive understanding of model tradeoffs across the major frontier reasoning and coding engines. You understand token economics, context degradation, reasoning token budgets, and fine-grained retrieval architectures.
- Uncompromising Tooling & Environment Opinions: You have an intentionally designed, battle-tested development workplace. Whether using advanced IDE setups (Cursor, Claude Code, customized NeoVim harnesses) or terminal-first workflows, you know precisely how to feed repository context, enforce schema types, and minimize hallucination loops.
- Test-First Obsession: You recognize that the bottleneck in modern software delivery is no longer code generation—it is verification and review. You maintain relentless automated test suites and deterministic CI/CD pipelines that prevent regressions before human eyes ever see a pull request.
- Systemic Problem Solver: You know when not to use an LLM. You draw sharp boundaries between deterministic software engineering principles and stochastic model behaviors.

The Litmus Test (Who This Role Is Not For)

We encourage you not to apply if:

- You view AI as an afterthought or a tab-completion plugin. If you haven't fundamentally altered how you scaffold projects, debug failure states, and run code review using LLMs over the past 18 months, this environment will not fit your working style.
- You cannot speak at length about frontier model architectures. You should be able to debate the real-world tradeoffs in reason

📌 Full Stack Software Engineer (Canada)
🏢 SpaceOps
📍 Canada