Applied AI Engineer, Internal Systems

Minerva is seeking an Applied AI Engineer to build and operate internal AI systems as a forward-deployed internal AI operations engineer. This role owns the company’s knowledge agent, automates core operating processes, and establishes guardrails while partnering closely with engineering and reporting to the COO.

Responsibilities

• Own the company’s knowledge agent, including ongoing operation and improvements.
• Automate the processes that run the company by translating workflows into reliable agent-powered systems.
• Run the skill library used across the agent ecosystem.
• Own the source of truth supporting internal operations and agent outcomes.
• Build the foundations under operating dashboards, including metrics, controls, and board reporting that are currently assembled manually from multiple systems.
• Put guardrails in place first to ensure safe operation of automated processes.
• Sit with your users to understand workflows and validate adoption.
• Measure adoption to track usage and impact across teams.

Required Qualifications

• Experience shipping agents or automations that coworkers use weekly, with clear evidence of usage and impact (for example, who stopped doing which tasks and how many hours were recovered).
• Production Python or TypeScript, with LLMs wired into real APIs.
• Comfort working with OAuth scopes, webhooks, and scheduled jobs, including understanding common failure modes.
• Ability to understand how a company runs and to communicate with stakeholders such as a recruiter, controller, and security auditor, improving the process before automating it.
• A controls mindset, including deciding approval flows and determining what happens when an agent is wrong.
• Up-to-date awareness of recent AI tooling shipped to production, including experience demonstrating your own setup.
• Clear writing that enables others to follow runbooks without needing direct prompting.
• Location in New York City or willingness to relocate.

Technologies

• Claude, MCP, Mastra, Temporal
• Slack, API
• Python, TypeScript
• OAuth scopes, webhooks, scheduled jobs
• Google Workspace, Notion, CRM, HRIS, ATS, HR, finance tools, compliance tools
• OAuth, LLMs

Role Context and Team Interface

• Minerva builds agents for marketers; this role builds agents for Minerva’s internal teams.
• Accountability sits with a forward-deployed engineer for Minerva itself, also described internally as an internal AI engineer or AI operations engineer.
• Primary customers include operations, finance, people, recruiting, sales, and the founders.
• Reporting line is to the COO, with close partnership with engineering (engineering owns anything that touches production).

What Success Looks Like

• First 90 days: the knowledge agent is live in Slack and used daily; onboarding for a new hire sets up access in every tool within an hour, with the access record writing automatically; at least three workflows run without a person in the loop and remain in use four weeks later.
• First 6 months: security questionnaires go out the same day without founder involvement; contract renewals, notice windows, and price increases surface before they cost the company money; each team has a system it would fight to keep.
• Ongoing metric: monthly hours of manual work retired, measured against a baseline established during the first month.

Compensation and Benefits

• Base salary: USD 211,000 to 236,000 per year, depending on experience.
• Meaningful equity: included as part of every full-time role.
• Medical, dental, and vision.
• Unlimited PTO.
• Monthly fitness stipend.
• Annual company offsite.

Company Location and Work Model

This role is onsite in New York, NY, working from the office at 300 Kent Avenue in Williamsburg, Brooklyn.

Our Current Stack

• Agents: Claude across the company (in Slack, through the API, with skills and MCP), plus Mastra for agents built in-house and Temporal for long-running work.
• Code and scheduling: Python and TypeScript, with scheduled jobs and webhooks.
• Systems your agents work across: Slack, Google Workspace, Notion, and company tooling including the CRM, recruiting, HR, finance, and compliance tools.

Additional Notes (Nice to Have)

• Experience rolling out Claude or another model platform to an entire company, including access, permissions, usage, and cost.
• Involvement in a SOC 2 or similar audit that produced evidence.
• Hands-on time with core startup systems such as HRIS, ATS, CRM, billing, and spend tools.

What This Role Is Not

• Not an IT helpdesk; the expectation is to automate tickets away rather than answer them.
• Not product engineering; customer-facing agents and production infrastructure belong to the engineering team.
• Not an AI strategy position; the focus is shipping systems people use.
• Not the owner of the work being automated; compliance and other owners retain responsibility for their areas while this role builds the systems that perform the work.

How to Apply

• Skip the cover letter.
• Include one example of something you built that other people used: a demo, a repository, or a few paragraphs describing what the process looked like before, what you built, how many people used it, and what broke.
• Applications without this example will not be reviewed.
• If you need an accommodation during the hiring process, email [email protected].

Work Authorization and Recruitment Fraud Policy

• Applicants must be authorized to work in the United States.
• Visa sponsorship is considered case by case; applicants should indicate needs on the application.
• Hiring runs through Ashby and @minerva.io email addresses.
• Minerva will not ask for money, bank details, or other financial information during the process.
• If an email appears to be from Minerva but looks wrong, forward it to [email protected].