Curiosity Tank is a consulting and education firm specializing in human-centered research, design development, and AI-informed product strategy, alongside hands-on learning programs. We conduct and teach design and user research to people and organizations around the world, including work at the intersection of human experience, emerging technologies, and AI systems.

Our solutions are affordable and repeatable. We apply our diverse research methods and traditional graphic design roots to deliver formidable results across all platforms, media, markets, and increasingly, AI-powered products and services.

Our consulting and training services cater to individuals and teams in the following areas:

• Design thinking and innovation workshop design and facilitation

• Qualitative research

• Systems thinking

• UX / interaction

• Prototyping

• Team, product, and service design creation and evolution

• UX, product, and education strategy

• Human-centered AI research, trust, and adoption

• Curricula development and execution

Let us deliver something remarkable for you. Or better yet, let us help you do it yourself.

Curiosity Tank is led by Michele Ronsen, a professional user researcher, designer, coach, and educator with 5,000+ hours of adult design and user research teaching expertise in academic institutions and 20+ years of corporate training, workshops, design, and research for Fortune 500s, startups, and frontier AI organizations.

We love questions! Learn more about us: www.CuriosityTank.com

Speaking engagements and other questions: hello@curiositytank.com

Clients include Cohere, Gusto, Stripe, Slack, Zillow, Invisalign, Xero, Microsoft, MileIQ, Autodesk, Square, Facebook, Kaiser Permanente, BNP Paribas, Best Buy, Blue Shield, PayPal, Mapfre, California College of the Arts, UC Berkeley, and many fantastic startups, non-profits, and mom and pop shops.

* Owned and led two global research pillars, LLM Safety & Alignment and Multilingual Language Accuracy & Generalization, informing model behavior, evaluation priorities, and long-term capability direction.

* Led multi-market research across the U.S., Canada, Japan, Korea, and the Middle East, advancing understanding of in-language and in-culture expectations for LLM behavior, trust, transparency, and failure handling. Designed and executed large-scale generative and evaluative studies in English, Arabic, Japanese, and Korean, examining language accuracy, formatting and presentation, safety behavior, refusals, and transparency across models.

* Translated findings into actionable guidance for model training, fine-tuning, evaluation, and safety design, influencing both near and long-term improvements and priorities. Insights justified roadmap, investment, and model iteration decisions.

* Outcomes directly informed the design of a net-new customizable safety guardrail framework, including response policies, transparency behaviors, configuration models, onboarding logic, and evaluation strategy. Work underpinned the development of a new Safety model (in progress) across controlled and public environments.

* In parallel, owned the multilingual research strategy supporting Arabic, Japanese, and Korean language performance. Led foundational research with senior LLM selectors and deployers across Modern Standard Arabic (MSA), Japanese, and Korean, identifying gaps in linguistic accuracy, cultural interpretation, formatting, and task execution across models. Established repeatable multilingual evaluation methods used to guide model improvement and training priorities

* Provided research leadership across frontier AI initiatives, including doc generation, feedback and preferences, system nomenclature, and evaluation practices. Led vendor strategy and RFPs for specialized research partners, scaling research capacity while improving rigor, consistency, and cost efficiency.

Design, develop and deliver engaging Design Thinking and User Research course curricula, presentations, handouts and workshop activities to professionals at all levels