Who You Are The Software Developer (Applied AI) is an established, productive developer who builds and runs
AI systems as software: the services, agents and integrations that turn DIRTT's knowledge and systems into AI capabilities for the business, for the Software Development department, for DIRTT's customers and partners. Those systems include Design Editor, the DIRTT platform used to design, configure and specify solutions, and the product catalogue behind it, as well as autonomous and semi-autonomous agents that carry out work over long task horizons.
Every DIRTT developer uses AI to write software. This role is different:
AI is the product. Working within the approach set by the technical lead, the Intermediate Software Developer (Applied AI) delivers AI capabilities that are evaluated, secure, cost-controlled and reliable in production, and grows toward independent ownership of them. The field moves quickly, so the role also tracks new models and capabilities and turns them into practical improvements.
Applied AI is the
primary focus of the role, not its only work. The Software Developer (Applied AI) is a member of the Platform team and also takes on Design Editor platform and cloud development as the team's priorities require.
What You'll Do AI Systems Development - Build AI services and integrations: applications on large language models (for example the Claude API and AWS Bedrock, including Bedrock Agents), with tool use, agents and multi-step workflows.
- Expose DIRTT's systems and knowledge to AI: build and curate DIRTT's knowledge bases, and build and maintain Model Context Protocol (MCP) servers and APIs over Design Editor, the product catalogue and those knowledge bases, with retrieval over DIRTT documents and data.
- Evaluate and harden AI systems: build evaluations, test sets, guardrails and human-in-the-loop checkpoints; measure quality, cost and latency; follow the data privacy and security practices for model use.
- Evaluate new models and capabilities as they are released: run them against DIRTT's evaluation suites, prototype what they make newly possible, and recommend when to adopt them.
- Help take AI capabilities from prototype to production for business initiatives, such as AI-assisted design in Design Editor; AI-driven automation of business processes such as order engineering; AI that helps product owners and analysts draft, check and complete requirements, so work is ready before development starts; and AI that sorts and routes customer support issues to the right team, with a first diagnosis.
- Contribute to the shared AI components, patterns and documentation that other developers build on.
Autonomous Agents & Agentic Tooling - Build autonomous and semi-autonomous agents that carry out multi-step work over long task horizons, with checkpoints, recovery from failure, cost limits, and clear points where a person approves or takes over.
- Build and improve agent harnesses and orchestration: agent loops, subagents and multi-agent workflows, including agents that carry out development work.
- Engineer context and memory: decide what goes into an agent's context, and how agents store, recall and prune knowledge across sessions and long-running tasks.
- Build agentic development tooling for Software Development: Claude Code skills, plugins, subagents and hooks that other developers use every day.
Platform Development - As a member of the Platform team, build and maintain Design Editor platform capabilities and cloud services as the team's priorities require, bringing AI into that work where it adds value.
- Take part in the team's support and maintenance of the platform.
Delivery & Ownership - Own your work end to end, from design through to production, with guidance from the technical lead on approach and priorities.
- Participate actively in the team's planning and delivery cadence in a scaled agile framework, and help keep it on track.
Engineering Craft & Quality - Deliver clean, well-tested work, and be able to explain and defend what you ship.
- Raise the team's quality through constructive code review, and recommend improvements to established patterns.
- Design for security and privacy from the start.
Cloud Operations - Deploy and operate AI services on AWS (serverless services, APIs and storage), with infrastructure as code and automated build, test and release through Jenkins or GitHub Actions.
- Make AI services and agents observable (logging, metrics, tracing of agent runs, quality and cost monitoring) and take part in incident response.
Collaboration, Influence & Mentorship - Work with business stakeholders, product owners and UX to understand needs and shape AI solutions.
- Explain the capabilities, limits and trade-offs of AI options clearly to the team and stakeholders.
- Support other developers as they start building on DIRTT's AI services.
What You'll Bring Required Qualifications - Bachelor's degree in Computer Science, Software Engineering, a related field, or equivalent experience.
- 3+ years of professional software development experience.
- At least 1 year of hands-on experience building applications on large language models, ideally with at least one in production: prompt and tool design, agents, retrieval, and evaluation.
- A solid understanding of how LLMs and agents work: context windows, tool calling, model strengths and failure modes, agent loops and harnesses, orchestration, and context and memory engineering.
- Proficiency in Java or Kotlin, and in Python.
- Experience building services on a major cloud platform, ideally AWS: serverless (Lambda, API Gateway), storage (S3), identity and access management.
- Awareness of the security, privacy and cost risks of AI systems, including prompt injection, data leakage through tools, and runaway agent cost.
- Keeps current with new models and AI tooling, and judges new ideas by evidence rather than hype.
- Testing discipline; version control (Git) and CI/CD.
- Clear communication with technical and business audiences.
Preferred Qualifications - Experience building Model Context Protocol (MCP) servers or similar tool-integration interfaces.
- Experience with the Claude API or AWS Bedrock, including Bedrock Agents.
- Experience building long-running or autonomous agents in production.
- Experience with agentic coding tools such as Claude Code, including writing skills, plugins or hooks.
- Experience with LLM evaluation frameworks and observability tooling.
- Infrastructure as code with Terraform; CI/CD with Jenkins or GitHub Actions.
- Familiarity with CAD, BIM, configure-to-order or manufacturing software.
What's In It for You - A competitive base salary plus a variable pay (bonus) program.
- Eligibility to participate in DIRTT's Employee Share Purchase Plan.
- Comprehensive health benefits, including medical, dental, and vision coverage (coverage and eligibility may vary by location).
- For in-office roles: continental breakfast, coffee, tea, soda, daily hot cafeteria lunch, and snacks to keep you energized.
- Short-term and long-term disability coverage, plus life, critical illness, and accidental death and dismemberment (AD&D) insurance.
- Mental health support, including access to an Employee and Family Assistance Program (EFAP).
- A thorough and supportive onboarding program designed to set you up for success.
