AI Agent Engineer – Commercial AI Transformation

Please note: This is a fully on-site position for the first six months, requiring attendance in the office five days per week.

Overview

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. We've already proven the model with a handful of production agents (including tools that generate executive briefs, estimate costs, and calculate customer value). 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. Knowing when to apply which is part of the job.

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. Comfort with that kind of environment matters as much as agent-building skill itself.

Key Responsibilities

Process Automation & Agent Building

• 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

Pipeline & Integration Work

• 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

Security & Governance Awareness

• 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

Engineering Excellence

• 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

Commercial Fluency & Collaboration

• 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

Success in This Role Looks Like

• A growing, reliable portfolio of agents in production, each automating a real, meaningful volume of work
• The team's agent-building work has real version control, testing, and release discipline behind it, not just working prototypes
• PoCs still move at PoC speed. Rigor shows up where it matters (production) not everywhere
• Commercial stakeholders trust that agents reflect how their work actually happens
• Pipelines and agents respect existing access control and governance boundaries without needing to be told twice

Required Qualifications

• 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

Preferred Qualifications

• 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