Native AI Engineering Lead

Focus Areas

1. Native AI Engineering Practice

• Define and operationalize how Aperium uses agentic coding tools (Codex, Claude Code, and successors) across the engineering lifecycle: specification, generation, review, refactoring, test authoring, and maintenance.
• Establish the guardrails, evaluation harnesses, and review gates that make AI-generated code safe to ship in compliance-driven environments, including deterministic build and test discipline (temperature 0, no hardcoded values, environment-variable configuration only), and build internal tooling and agent workflows that raise throughput and reliability across the broader engineering organization.

2. Ontology and Knowledge Modeling

• Own the shared ontology that grounds agent reasoning across Aperium's domains. Define entities, relationships, and constraints, and the mapping between the ontology and the underlying systems of record.
• Build an engineering ontology that absorbs the company's own software-development knowledge into a unified semantic layer: Jira issues, epics, and workflows; Confluence pages and runbooks; architecture decision records and design documents; and the code, services, and repositories they describe. Resolve these fragmented sources into coherent entities (services, components, tickets, decisions, owners, dependencies) and the relationships between them.
• Use that semantic layer to ground agentic development so Codex and Claude Code reason against an accurate model of the system: which service a ticket touches, which decision record governs a component, who owns it, and what depends on it, rather than re-deriving context from scattered free text on every task.
• Ensure agents resolve references, infer intent, and act consistently with the domain model rather than free-form text alone, and integrate the ontology with Aperium's memory, retrieval, and MCP tool layer. Keep it continuously synchronized with source systems so it stays current and auditable across the engineering and product surfaces.

3. Systems Architecture and Delivery

• Architect and ship the Native AI group's systems end to end, from data and ontology layers through agent orchestration and the MCP tool surface.
• Set engineering standards for the group consistent with company-wide development guidelines, including branch and commit discipline against the shared repository and Git Flow conventions.
• Make build-versus-integrate decisions, manage technical risk, and keep the roadmap aligned with platform priorities and the competitive model landscape.

4. Team Leadership and Growth

• Recruit, hire, and develop the founding engineers of the Native AI group; define its operating model, rituals, and quality bar; and mentor engineers on agentic development technique and ontology-driven design to raise AI-engineering capability across the wider organization.
• Represent the group's work to senior leadership and partners, and translate frontier capability into a credible, executable roadmap.

Qualifications

• 10+ years of software engineering experience, including time at staff, principal, or equivalent senior IC level, with a track record of shipping production systems at scale.
• Demonstrated leadership scope: has led teams or set technical direction for an engineering function, or operated as a founding or head-of-engineering-level technologist. A hands-on leader who still writes code.
• Deep, current, hands-on fluency with agentic coding tools, specifically Codex and/or Claude Code, used as core engineering infrastructure rather than occasional assistance. Can speak concretely to workflows, failure modes, evaluation, and the limits of these tools.
• Strong working understanding of ontology and knowledge modeling: entity-relationship modeling, taxonomies, knowledge graphs, and grounding agent behavior in a formal domain model. Experience unifying fragmented engineering knowledge (Jira, Confluence, architecture docs, code) into a coherent semantic layer is a strong plus; familiarity with graph stores and retrieval architectures is expected.
• Excellent Python engineering discipline and systems design judgment; experience with LLM application architecture, tool-calling, retrieval, and evaluation.
• Familiarity with the Model Context Protocol (MCP) as an integration primitive, or the ability to learn it quickly and reason about it as an architectural foundation. Experience in compliance-aware or high-assurance environments is a plus; willingness to work within ITAR-aware and CMMC-aligned constraints where applicable.
• BS, MS, or PhD in Computer Science, Mathematics, or a related field, or equivalent demonstrated expertise.

Key Competencies

Agentic coding tools (Codex / Claude Code)

Ontology and knowledge modeling

LLM application and agent architecture

MCP and tool-calling design

Hands-on systems architecture

Engineering leadership and team building

Evaluation and guardrails for AI systems

Python engineering discipline