Principal ML & AI Engineer, ApexIQ

Principal ML & AI Engineer, ApexIQ

About the role

Mimecast is reimagining how security and compliance teams work. Instead of jumping between consoles, teams will have one place to see their organization's risk, understand it in plain language, and work alongside AI agents that get the work done. Each agent does one job well, such as reviewing configuration, investigating threats or briefing a leader on risk. Each one shows its reasoning, offers clear choices, and changes nothing without the admin's approval. Security teams have too many tools and too few people, and this is a new way to work.

We're hiring a Principal ML & AI Engineer to build ApexIQ, the intelligence layer that powers the Mimecast Apex Platform. ApexIQ is the shared knowledge, reasoning, memory and evaluation that every agent runs on. You'll set its technical direction from day one, and your work will decide what a security platform built around agents looks like. It's a rare chance to define the category rather than follow it. You'll work closely with product, engineering and data science, and you'll set the standard for how Mimecast builds, measures and improves agents across the portfolio.

What you'll do

• Architect, build and run ApexIQ, the layer every agent shares: knowledge retrieval, reasoning, tool use, memory and context about each customer's environment.

• Ship agents to production on top of ApexIQ that give admins plain answers and concrete options.

• Unify product documentation, recommended defaults, live configuration and threat signals into a knowledge foundation that agents can query and cite.

• Turn deterministic rules engines and reports into agent workflows that explain findings, propose fixes, record exceptions and recheck results.

• Build the evaluation framework for agent quality: offline test sets, LLM-as-judge and human review, and live monitoring for accuracy, groundedness and safety.

• Instrument agents end to end, including tool calls and MCP server traces, so every answer can be explained and every failure diagnosed.

• Set guardrails for agent actions: approval steps, risk tiers, undo, and audit trails.

• Work with product to choose which workflows to automate next, based on customer impact and data readiness.

• Try new approaches in agent design, memory and multi-agent coordination, and turn the ones that work into reusable patterns in ApexIQ.

• Mentor engineers and data scientists, and help shape AI standards across product teams.

What you'll bring

• 8+ years in ML engineering, data science or applied AI, including 2+ years shipping LLM-based products to production.

• Experience designing shared ML or LLM platforms that several products or teams build on.

• Hands-on experience building agents or multi-step LLM workflows: tool use, retrieval, planning and memory.

• A strong track record of designing evaluations for generative systems and using the results to improve quality.

• Fluency in Python and modern LLM tooling. Experience with AWS and hosted model services such as Amazon Bedrock.

• Experience with observability for ML or LLM systems, such as tracing, eval pipelines and drift monitoring.

• The ability to explain technical tradeoffs in plain language to product, sales and customer teams.

• A product mindset: you care whether the agent helps the person using it, not just whether the model scores well.

Nice to have

• Background in email security, cybersecurity, or governance and compliance data.

• Experience with Model Context Protocol (MCP) servers or similar agent tool frameworks.

• Experience building human-in-the-loop systems where an admin reviews and approves actions.

• Published work, open source contributions or talks on agents, evaluation or applied LLMs.

- The base salary range for this position is $172,000 − $258,000 plus benefits. This range represents the minimum and maximum new hire compensation for this role. The position may also be eligible for incentive plans and additional benefits, in accordance with company policy and local regulations. Our salary ranges are determined by role, level, and location with individual compensation also dependent on factors such as qualifications, experience, and skills. Final offers will reflect these considerations and may vary accordingly.

Belonging at Mimecast

Cybersecurity is a community effort. That’s why we’re committed to building an inclusive, diverse community that celebrates and welcomes everyone – unless they’re a cybercriminal, of course.

We’re proud to be an Equal Opportunity and Affirmative Action Employer, and we’d encourage you to join us whatever your background. We particularly welcome applicants from traditionally underrepresented groups.

We consider everyone equally: your race, age, religion, sexual orientation, gender identity, ability, marital status, nationality, or any other protected characteristic won’t affect your application.

If you require any adjustments or accommodations due to a disability, or any other reason that may help you in your interview process, please let us know by emailing [email protected].

Due to certain obligations to our customers, an offer of employment will be subject to your successful completion of applicable background checks, conducted in accordance with local law.

It is unlawful in Massachusetts to require or administer a lie detector test as a condition of employment or continued employment.