Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI Operations & Training Project Manager with 2+ years leading RLHF, model evaluations, and large-scale annotation programs at Cohere, micro1, and Invisible Technologies. Proven track record managing distributed teams of 300+ data trainers, improving annotation accuracy by 30%, and reducing QA defects by 30% through systematic workflow design and training development. Expertise in translating complex AI alignment requirements into structured processes, evaluation rubrics, and quality frameworks. Specialized in RLHF operations, prompt engineering coordination, video annotation management, and AI model evaluation programs.…AI Operations & Training Project Manager with 2+ years leading RLHF, model evaluations, and large-scale annotation programs at Cohere, micro1, and Invisible Technologies. Proven track record managing distributed teams of 300+ data trainers, improving annotation accuracy by 30%, and reducing QA defects by 30% through systematic workflow design and training development. Expertise in translating complex AI alignment requirements into structured processes, evaluation rubrics, and quality frameworks. Specialized in RLHF operations, prompt engineering coordination, video annotation management, and AI model evaluation programs.

Ben Javinsky

Project Manager, AI Strategy Consultant, AI Prompt Engineer





AI Operations & Training Project Manager with 2+ years leading RLHF, model evaluations, and large-scale annotation programs at Cohere, micro1, and Invisible Technologies. Proven track record managing distributed teams of 300+ data trainers, improving annotation accuracy by 30%, and reducing QA defects by 30% through systematic workflow design and training development. Expertise in translating complex AI alignment requirements into structured processes, evaluation rubrics, and quality frameworks. Specialized in RLHF operations, prompt engineering coordination, video annotation management, and AI model evaluation programs.…AI Operations & Training Project Manager with 2+ years leading RLHF, model evaluations, and large-scale annotation programs at Cohere, micro1, and Invisible Technologies. Proven track record managing distributed teams of 300+ data trainers, improving annotation accuracy by 30%, and reducing QA defects by 30% through systematic workflow design and training development. Expertise in translating complex AI alignment requirements into structured processes, evaluation rubrics, and quality frameworks. Specialized in RLHF operations, prompt engineering coordination, video annotation management, and AI model evaluation programs.

Available to hire

AI Operations & Training Project Manager with 2+ years leading RLHF, model evaluations, and large-scale annotation programs at Cohere, micro1, and Invisible Technologies. Proven track record managing distributed teams of 300+ data trainers, improving annotation accuracy by 30%, and reducing QA defects by 30% through systematic workflow design and training development. Expertise in translating complex AI alignment requirements into structured processes, evaluation rubrics, and quality frameworks. Specialized in RLHF operations, prompt engineering coordination, video annotation management, and AI model evaluation programs.

Skills

Experience Level

Expert

Expert

Expert

Intermediate

Intermediate

Beginner

Language

English

Fluent

Work Experience

AI Data Trainer at Cohere

October 13, 2025 - Present

• Perform RLHF (Reinforcement Learning from Human Feedback) annotation tasks for Cohere's language model training programs, providing preference feedback and response rankings. • Follow structured annotation guidelines to ensure consistency in training data quality across prompt engineering, conversational AI, and code generation tasks. • Review and evaluate AI-generated outputs for accuracy, safety, and helpfulness according to established evaluation criteria. • Provide structured feedback on model responses to support iterative model improvement and alignment objectives. • Maintain clear documentation of annotation decisions and surface edge cases or unclear requirements to support cross-functional alignment.

Human Data Manager (AI Training Operations) at micro1

June 16, 2025 - August 15, 2025

• Led end-to-end quality assurance operations for AI training programs, reducing annotation defects by 30% through systematic process redesign. • Managed distributed teams of data annotators for a prompt engineering project. • Developed comprehensive training materials, evaluation rubrics, and performance frameworks ensuring consistency across high-complexity tasks. • Improved operational efficiency by 20% through targeted workflow optimization and capability-building interventions. • Translated stakeholder and SME requirements into actionable training documentation and quality standards. • Facilitated remote coordination across project leads, ensuring predictable delivery timelines and alignment on success metrics.

Project Manager at Invisible Technologies

June 10, 2024 - June 13, 2025

• Co-led operations for 300+ distributed data trainers delivering large-scale RLHF and model evaluation programs for enterprise AI clients. • Improved annotation accuracy by 30% through development of structured quality systems, feedback loops, and training frameworks. • Built centralized documentation hub (single source of truth) improving stakeholder visibility and reducing onboarding time by 40%. • Designed and facilitated recurring knowledge-sharing sessions, translating complex technical requirements into executable workflows. • Implemented coaching and performance management strategies, elevating team capability across prompt engineering and evaluation tasks. • Identified process gaps and drove continuous improvement initiatives, increasing throughput while maintaining quality standards. • Partnered with engineering and research teams to refine annotation schemas and evaluation criteria based on model performance data.

Advanced AI Data Trainer & Quality Analyst at Invisible Technologies

April 8, 2024 - June 7, 2024

• Maintained 90%+ accuracy across RLHF and evaluation tasks for leading AI model training initiatives. • Developed onboarding curriculum, QA training workshops, and written guidelines accelerating new trainer ramp-up time by 50%. • Conducted systematic quality reviews of trainer outputs, identifying error patterns and improvement opportunities across diverse use cases. • Synthesized quality findings into actionable feedback informing training rubrics, annotation schemas, and model alignment strategies. • Collaborated with cross-functional partners (engineering, research, product) to resolve edge cases and maintain predictable project throughput. • Led pilot testing of new annotation frameworks, providing structured feedback to refine task design before full team rollout.