Toloka AI supports frontier model post-training by building domain-specific reinforcement learning environments, tasks, and evaluation frameworks designed by real practitioners.
Mindrift, powered by Toloka — a leading enterprise AI and machine learning data partner since 2014 — connects top domain experts with cutting-edge AI initiatives. Backed by Toloka’s deep expertise in scalable data generation, crowd technology, and applied ML systems, Mindrift enables experts to shape how next-generation generative models learn, reason, and perform.
To do this credibly, we are building a team of McKinsey, BCG, and Bain consultants who can convert authentic project experience into end-to-end examples — from problem structuring and work planning to analysis, synthesis, and client-ready recommendations.
This requirement is non-negotiable and helps us ensure the domain is built by practitioners with firsthand MBB training and standards.
You should have hands-on project experience:
Build realistic consulting project environments : create detailed project scenarios (industry context, financials, constraints, incomplete information).
Design structured consulting tasks for AI Agents : break projects into tasks that mirror real consulting work such as market sizing, due diligence, cost reduction, growth strategy, or operational diagnosis.
Define evaluation criteria and quality standards : develop a grading approach, evaluation criteria and golden solutions for each task that will be used to train and calibrate an LLM-based grading system that evaluates AI outputs at scale.
This is a remote individual-contributor project-based role focused on analytical design and evaluation.
Clear written English (B2+)