AI Agent Engineer – Commercial AI Transformation

Diligent Corporation•Vancouver, BC

11d•CA$120,000 - CA$150,000•Hybrid

About The Position

Diligent's Commercial AI Transformation function builds AI agents that automate real workflows across our commercial organization, from sales and customer success to internal operations. This role exists to grow that portfolio, working alongside a teammate who also designs and builds agents. You'll bring two things the team needs more of: a track record of automating real business processes at volume, and genuine software engineering discipline (version control, testing, release practices) that the team can build on as it scales. At the same time, you need the agility to move fast on a proof of concept without forcing full engineering process onto something that's still proving itself out. You'll also step into a real enterprise integration environment on day one, pulling from source systems, processing through a data warehouse, feeding an enterprise search/AI index, and operating inside access control and governance boundaries that are already in place.

Requirements

Bachelor's degree in Computer Science, AI/ML, or a related field
4+ years building software, including demonstrated experience automating business processes at meaningful scale (not one-off scripts)
Strong software development lifecycle (SDLC) fundamentals: version control (Git), code review practices, testing, and release/deployment discipline
Hands-on experience building integrations or data pipelines using an iPaaS/automation platform (e.g., Workato, Boomi, Mulesoft, or similar)
Experience working with a cloud data warehouse (e.g., Snowflake) for data ingestion, transformation, or processing
Recent hands-on experience designing, building, and deploying LLM-based agents or agentic workflows into production
Working understanding of enterprise identity and access concepts (SSO, OAuth, group-based permissions) and how they constrain what a pipeline or agent can access
Demonstrated judgment about when to move fast and informal (PoC stage) versus when to apply full engineering rigor (production stage)
Genuine interest in and some exposure to how a commercial org (sales, CS, or ops) functions

Nice To Haves

Experience with enterprise search or knowledge platforms (e.g., Glean) and how they scope and surface indexed content
Familiarity with Microsoft 365 ecosystem tooling relevant to data governance (e.g., Purview, Defender, Graph API)
Experience working with SIEM or logging platforms (e.g., Panther, Splunk) from an integration or engineering standpoint
Experience in a presales, customer success, or commercial operations environment
Familiarity with Salesforce or similar commercial data systems
Experience mentoring or upskilling less traditionally-trained engineers on SDLC best practices
Test-Driven Development experience

Responsibilities

Map business processes independently when needed, and design, build, and deploy AI agents and agent chains that automate them
Refine agents through iteration: tightening prompts, handling edge cases, improving reliability based on real usage
Move quickly through early-stage PoCs, then apply appropriate engineering rigor once an agent is heading toward production
Build and maintain data pipelines that pull from source systems (e.g., Microsoft Graph API, Teams, Snowflake) into a data warehouse, applying appropriate filtering, summarization, and sensitivity handling before anything is indexed or surfaced
Work within an iPaaS/integration platform (e.g., Workato) to build and maintain recipes and API endpoints that connect systems together, including logging and monitoring for those integrations
Understand how enterprise search/AI indexing tools (e.g., Glean) consume processed data, including index scoping, access restrictions by group, and how retrieval respects underlying permissions
Apply access control patterns correctly: privileged access boundaries, IP whitelisting, OAuth-based endpoint protection, and group-based restrictions on what data or tools a user can reach
Understand how identity and access (e.g., Okta/SSO) and logging/SIEM tooling (e.g., Panther) fit around the systems you're building, enough to build in a way that doesn't create gaps
Work with sensitivity tagging and data minimization principles when pulling raw data (e.g., removing what isn't needed, redacting or filtering employee-specific content) before it moves further into the pipeline
Introduce and drive adoption of solid software engineering practices across the team's agent-building work: version control, code review discipline, testing, and release/deployment practices
Set a practical bar for what "production-grade" means for an agent, distinct from what's acceptable in a fast-moving PoC, and help the team recognize which stage something is in
Own agents from prototype through production-grade deployment, including error handling, monitoring, and failure-mode recovery
Extend and reuse existing shared infrastructure rather than duplicating capability; apply an "extend, don't rebuild" discipline
Understand enough about how sales, customer success, and commercial operations actually work to design agents that reflect reality, not a theoretical process
Partner closely with the teammate who also designs and builds agents, sharing the design and build workload flexibly
Bring engineering best practices to the team without slowing down the team's pace on early-stage work