โจ I shall divine your name through the ancient arts! โจ
โจ Help train the spirits by sharing your real name (optional):
This helps the spirits learn and improve their predictions!
โจ My crystal ball never lies! โจ
The story behind the crystal ball
This interactive name-guessing experience combines cutting-edge web development with extensive cultural research and our own custom-built AI-powered predictions. Built with vanilla JavaScript and TensorFlow.js, it demonstrates sophisticated algorithms, cultural sensitivity, and modern UX design while remaining fun and accessible. All AI/ML processing is done locally - no data is sent to ChatGPT, OpenAI, or any commercial AI services.
Charlotte Larson Freeman - Software Engineer (New Grad, March 2026)
Passionate about human-centered, accessible AI and creating engaging user experiences. Currently working as an R&D Intern at Silfab Solar Inc., developing image processing software for solar cell analysis.
This is a fun, educational project only! ๐
Your data is completely private and secure:
How it works: When you provide feedback (correct/wrong guesses or your real name), this data is stored locally in your browser. Our custom TensorFlow.js machine learning model then trains on this local data to personalize predictions for you. The model itself is not saved - it only exists in memory during your session. All AI processing happens locally - no data is sent to ChatGPT, OpenAI, or any commercial AI services. Additionally, training data (quiz answers + success/failure + your first name if you provide it) may be sent to GitHub Issues for global model training. Your first name (if provided) is the most valuable data for training - it helps the model learn which names correspond to which characteristics. This data will be visible in public GitHub Issues. Providing your name is completely optional.
Discover the mystical science behind our hybrid AI-powered name divination
Our crystal ball draws from extensive research and massive databases:
Our own implementation combines rule-based logic with machine learning. All AI/ML processing is done locally - no data is sent to ChatGPT, OpenAI, or any commercial AI services.
Our advanced system combines rule-based logic with machine learning:
Our algorithm analyzes 15+ dimensions for each name based on your responses:
Note: This is our own TensorFlow.js implementation. No data is sent to ChatGPT, OpenAI, or any commercial AI services.
Our system learns and improves with every interaction:
The hybrid system generates your personalized name predictions using sophisticated scoring:
SVG-based continent and country selection with 200+ clickable regions. Dynamic styling, hover effects, and responsive design for seamless geographic input.
Respectfully researched cultural/linguistic groups spanning European, Asian, African, Middle Eastern, Indigenous, and more traditions.
Pre-built indexes for 15+ dimensions including cultural origin, semantic meanings, personality traits, traditional significance, and socioeconomic associations.
Our own implementation combining sophisticated rule-based scoring with TensorFlow.js neural networks. 70% algorithmic precision + 30% machine learning adaptability. All processing is local - no commercial AI services are used.
350+ curated gender-neutral names from reputable baby name websites. Inclusive approach that respects non-binary identities with research-backed selections.
Every name category backed by reputable sources: SSA data, academic studies, mommy blogs, baby name websites, genealogical records, and cultural research.
All data stays on your device. No tracking, no analytics, no external servers. Feedback forms optional and transparent about data handling.
Our custom local TensorFlow.js model improves with each quiz completion and user feedback, adapting to real-world naming patterns while respecting privacy. No data is sent to ChatGPT, OpenAI, or any commercial AI services.
Optimized algorithm processes 47,689 names across 15+ dimensions in milliseconds. Efficient filtering and scoring deliver instant results.
Groundbreaking study revealing that 84% of blue state names are non-English, while 90% of red state names are English. Political identity is the strongest predictor of name choice.
Read NPR Study โAuthoritative research by the creator of "The Baby Name Wizard" revealing that rural white America is the LEAST traditional in naming patterns - a counterintuitive finding.
Explore Namerology โOfficial baby name data from 1880-2024, including popularity rankings and birth counts by state. The gold standard for name statistics.
View Official Data โPeer-reviewed studies showing how family income, education level, and cultural background influence naming choices across different social strata.
Academic Study โPsychological studies demonstrating that people prefer names starting with letters from their own name - a phenomenon called the "name letter effect."
Psychology Research โResearch on naming patterns across different cultural and religious traditions, including naming ceremonies and their significance.
Cultural Studies โCurated database of 350+ gender-neutral names from reputable baby name websites, mommy blogs, and SSA statistical analysis of names with balanced male/female usage.
Nameberry Research โ40+ cultural/linguistic groups researched through genealogical records, government naming databases, academic surveys, and reputable cultural naming resources worldwide.
Global Name Usage โOur algorithm employs 15+ weighted factors based on research and extensive cultural databases. Each dimension contributes points to calculate the best-matching names:
40+ Cultural Groups: Maps selected countries/continents to specific cultural/linguistic traditions. Respects diverse naming patterns across European, Asian, African, Middle Eastern, Indigenous, and other cultures.
NPR Research: Strongest predictor - traditional values correlate with English/classical names, progressive values with diverse cultural names.
NPR Finding: 84% of blue state names are non-English, 90% of red state names are English. Matches language preferences to cultural identity.
Multi-Faith Analysis: Christian, Jewish, Muslim, Hindu, Buddhist, and non-religious naming associations drawn from religious texts and cultural research.
300+ Historical Names: Ancient, biblical, classical, royal names vs modern/trendy selections. Matches family tradition importance.
500+ Categorized Names: Semantic associations including water, fire, earth, nature, moon, sun, strength, wisdom, peace, music, nobility, life, and love.
Namerology Research: Counterintuitive finding - rural areas prefer non-traditional names, urban areas prefer classical names.
300+ Names Analyzed: How names are perceived (elegant, strong, friendly, intelligent, creative, unique, traditional, modern, natural).
300+ Names Categorized: Names that convey authority, strength, warmth, intelligence, creativity, uniqueness, tradition, nature connection, trust.
Academic Research: Elite/aristocratic vs working-class naming patterns. 150+ names categorized across socioeconomic spectrum.
150+ Names Mapped: Coastal/urban vs rural vs suburban vs small town vs mountain preferences reflected in name choices.
Psychology Research: People unconsciously prefer names starting with letters from their own name - a documented psychological phenomenon.
Social Perception: How names are received by others - formal/professional, creative/artistic, friendly/approachable, internationally recognized.
Cultural Studies: Openness to diverse name origins correlates with selection of international vs traditional English names.
Linguistic Analysis: Length preferences, vowel/consonant patterns, sound aesthetics matched to user preferences.
SSA State Data: All 50 states plus territories. Names filtered and scored based on birth state popularity and regional trends.
* Results are for entertainment purposes. Name prediction accuracy varies based on available research data and user responses. Algorithm based on peer-reviewed studies and authoritative research.
Connect with the developer behind the magic
Software Engineer
BS in Computer Science, Western Washington University - March 2026
Silfab Solar Inc. | June 2025 - Present
Developing image processing software for solar cell analysis, using AI-assisted development tools while maintaining full QA responsibility.
Western Washington University | January 2025 - Present
Researching ethical implications of AI and ableism in LLMs, co-developing coding schemas for bias analysis.
Java, Python, C, JavaScript, HTML/CSS, Racket, SQL, React
Git, IntelliJ, Visual Studio, ImageJ, LaTeX, Bash, OpenCV
Image Analysis, AI Ethics, Human-Centered Design, Data Visualization
Share your feedback, suggestions, or bug reports!