Lead AI/ML Engineer: Scale Production Systems

Ciklum; United States; On-site; USD 180,000 - 280,000; Full time

Benefits offered by this job

Healthcare

Life Insurance

Disability insurance

Udemy access

Certifications

Job summary

Ciklum is seeking a Lead Artificial Intelligence/ML Engineer to drive development in the US, overseeing a cross-functional, fast-moving team of engineers and contractors. You will shape architecture reviews, evaluation strategies, and production readiness to deliver scalable AI/ML solutions.

The role emphasizes hands-on oversight of prompts, agent architectures, and eval suites, with a focus on flat, fast decision-making and empowering pods to ship impactful projects across diverse domains.

Qualifications

• 8+ years engineering experience.
• 3+ years managing engineers who ship AI/ML or LLM systems in production.
• Hands-on enough to review agent architectures, prompts, and eval suites.
• Operates in a flat, fast structure—player-coach style.

Responsibilities

• Manage engineers across the central platform team and delivery pods (contractors and internal engineers).
• Own engineering standards: code and agent quality, review practices, definition of done, production readiness.
• Own contractor performance and the conversion pipeline with the sourcing partner.
• Partner with the platform lead on templates, pipeline, evals—ensure every pod uses them.
• Unblock delivery daily; keep pods shipping without becoming a gatekeeper.

Skills

Team leadership

AI/ML systems

Contractor management

Code review

Job description

Ciklum is seeking a Lead Artificial Intelligence/ML Engineer to drive development in the US, overseeing a cross-functional, fast-moving team of engineers and contractors. You will shape architecture reviews, evaluation strategies, and production readiness to deliver scalable AI/ML solutions.

The role emphasizes hands-on oversight of prompts, agent architectures, and eval suites, with a focus on flat, fast decision-making and empowering pods to ship impactful projects across diverse domains.