Skills & Expertise

Reinforcement Learning from Human Feedback — providing structured human preference signals to guide model alignment and improve response quality.

Assessing AI-generated content for accuracy, reasoning quality, and guideline adherence. Identifying improvement opportunities through structured evaluations.

Performing high-quality human annotations to identify factual inaccuracies, communication gaps, and adherence to system guidelines.

Applying fine-grained qualitative judgments and nuanced evaluation rubrics to compare multiple model outputs and rank responses.

Crafting and refining prompts to elicit high-quality, instruction-adherent responses from large language models.

Serving as the human oversight layer in AI pipelines to catch errors, ensure safety, and provide training signal for continuous model improvement.

Native-level French fluency used in government translation, healthcare interpretation, and evaluating French LLM outputs for reasoning and clarity.

Primary professional language. Expert in syntax, semantics, and cultural context — applied to structured content assessment and stakeholder communication.

Native Lingala speaker with deep cultural and contextual understanding, enabling nuanced cross-cultural communication.

Precise French-to-English interpretation and translation in government and healthcare contexts requiring exact meaning, consistency, and contextual accuracy.

Used for database design, data validation, batch-processing workflows, and translating complex datasets into clear analytical outputs.

Designing and maintaining databases, applying quality control, and conducting multi-source data reconciliation across structured datasets.

Creating visual analytical outputs using Tableau to support data-driven decision making and communicate findings to stakeholders.

Building and maintaining data pipelines for extraction, transformation, and loading to ensure metadata integrity and consistency.

Clear, precise written communication applied to evaluation reports, translation output, structured feedback, and compliance documentation.

Consistently catching subtle errors in terminology, reasoning chains, factual claims, and guideline adherence across high-stakes evaluation tasks.

Successfully pivoting across AI evaluation, government translation, healthcare interpretation, and data analysis roles with consistent high performance.

Structured approach to breaking down complex problems, comparing options, and producing reasoned, evidence-backed evaluations and assessments.