Lead AI Engineer

About Axya

Role overview

What you will own

• Own AI-layer architecture, including agent and LLM workflows, retrieval, tool use, and links to customer ERP and procurement data.
• Build evaluation frameworks, test sets, and monitoring to measure accuracy, reliability, and improvement.
• Take AI features from prototype to dependable, observable, cost-aware production systems.
• Select and benchmark models, vendors, and frameworks, balancing quality, latency, cost, and customer data constraints.
• Partner on enterprise data handling, access control, privacy, and auditability.
• Set AI engineering practices, mentor engineers, and turn customer usage and failures into improvements.

Must have

• Production experience shipping LLM-based or ML-powered systems to real users.
• Strong software engineering fundamentals, Python fluency, and experience building and operating backend services.
• Hands-on experience with prompting, tool use, retrieval, agent and workflow orchestration, and evaluation.
• Sound architecture judgment and ability to explain tradeoffs and own outcomes.
• Technical leadership through direction-setting, review, mentoring, or project leadership; formal reports are not required.
• Clear communication with technical and non-technical stakeholders.

Preferred experience

• Strongly preferred: enterprise or B2B products; structured ERP, supply-chain, procurement, finance, or operations data; evaluation and observability tooling; early-stage or high-growth environments; or model cost, latency, and selection tradeoffs at scale.
• Nice to have: fine-tuning or model adaptation; manufacturing, procurement, or supply-chain domain experience; familiarity with enterprise security and compliance; French fluency.
• No specific years of experience, degree, or employer background is required.

Success milestones

• Within 30 days, understand architecture, customer use cases, and AI failures; ship an improvement and assess key risks and opportunities.
• Within 90 days, establish a baseline evaluation framework used in release decisions, lead a major AI feature or re-architecture to production, and make quality, latency, and cost visible.
• Within 6 to 12 months, measurably improve accuracy and reliability, co-own an AI roadmap, and establish repeatable evaluation, release, and observability practices.

Benefits

• Unlimited vacation, health insurance, and a flexible work schedule.
• The employer describes the role as 100% remote.