résumé

Coursework in machine learning and cognitive science

CS229 Machine Learning; SYMSYS1 Minds and Machines; PSYCH80 Mind, Brain, and Behavior.

Semester abroad · GPA: 18.7/20

Coursework: Data Science, Machine Learning, Econometrics, Research Methods.

CSIE & FABIZ · Double bachelor’s degrees

B.Sc. in Computer Science and B.Sc. in Economics.

Research collaboration with Kazuki Kawamura

• Developed personalized LLM agents that use adaptive interviews to model social preferences and assess connections based on users’ goals and context.
• Built a Python/PyTorch recommendation engine that learns from sparse pairwise scores to rank candidates and construct degree-constrained social graphs.
• Designed a longitudinal field study comparing feedback-updated and interview-only agents on compatibility predictions for unseen people. Planned measures: agreement with human judgments, match quality, LLM query efficiency, and interest in reconnecting.

• Built a system to model individual behavior and relationship dynamics from WhatsApp conversation data.
• Generated behavioral profiles linked to supporting messages; incorporated user corrections and tracked changes in expressed emotions and communication patterns.
• Compared users’ stated preferences with observed conversational behavior and interaction feedback.

• Fine-tuned Qwen3.8-27B on SocSci210 with LoRA; compared standard SFT with additional objectives for matching response distributions within experimental conditions and predicting effects between conditions.
• Evaluated generalization to held-out studies using response-distribution and experimental-effect prediction errors; conducted ablations to isolate each additional loss term’s contribution.

Semantic Search Engine

• Built an end-to-end Python pipeline for natural-language search over 18,611 university student profiles: data processing, embeddings, FAISS/HNSW approximate nearest-neighbor search, cosine similarity, and top-k retrieval.

Software Engineer, Machine Learning · Tokyo, Japan

• Built and deployed the end-to-end Python ML backend as sole ML engineer for an app with 5,000+ users: outfit generation, personalized ranking, recommendation APIs, and SQL storage.
• Tuned behavioral feature weights with Taguchi orthogonal arrays; increased selection of recommended outfits by 37% versus the previous model in a production A/B test.

Data Analyst Intern · Brisbane, Australia

• Designed and deployed a Python/SQL test-analysis platform adopted as the team’s standard workflow: 10M+ sensor readings across 60+ motor/controller test batches. Cut data preparation and verification time by 83% (2 hours to 20 minutes per batch).
• Automated ingestion, validation, and sensor time alignment, detecting 85% of validated anomalies; delivered APIs, an interactive dashboard, reproducible reports, and optimized SQL for hardware comparisons and anomaly analysis.