Lead Optimization Scientist — Convex & LP Expert

EPAM Systems; United States; Remote; USD 140,000 - 220,000; Full time

Job summary

EPAM Systems is seeking a Lead Data Scientist to critically assess current optimization models and identify weaknesses. You will recommend mathematically sound improvements and document findings for stakeholders and technical teams.

You will apply convex, linear, and nonlinear programming methods, evaluating problem formulations and solver choices, with a focus on delivering robust, scalable solutions across domains.

Qualifications

• 5+ years of experience in optimization, operations research, or applied mathematics.
• In-depth knowledge of Convex Optimization, including convexity, duality, and optimality conditions.
• Strong command of Linear Programming, including formulation, the simplex and interior-point methods, sensitivity analysis, and duality.
• Very good understanding of Linear and Non-linear Programming, including their theory, solution methods, and limitations.
• English proficiency at B2 level or higher.
• Nice to have Background in Oil & Gas domain

Responsibilities

• Identify flags such as formulation errors, infeasibilities, unboundedness, poor scaling, non-convexities, and inappropriate solver or method choices
• Assess whether each problem is modeled with the right approach, whether linear, non-linear, or convex formulation
• Determine the suitability of linear versus non-linear programming methods for a given problem
• Suggest clear, actionable improvements to models, methods, and solution approaches
• Document recommended changes and findings for stakeholders and technical teams

Skills

Optimization

Convex optimization

Linear programming

Nonlinear programming

Duality

English (B2+)

Job description

EPAM Systems is seeking a Lead Data Scientist to critically assess current optimization models and identify weaknesses. You will recommend mathematically sound improvements and document findings for stakeholders and technical teams.

You will apply convex, linear, and nonlinear programming methods, evaluating problem formulations and solver choices, with a focus on delivering robust, scalable solutions across domains.