Arushi Ranjankumar Singh
Data Analyst who also builds Machine Learning Pipelines
New to Contra
Arushi Ranjankumar is building their profile!
Built a full-stack music analytics dashboard processing user listening history without Spotify API auth — accepts uploaded StreamingHistory.json exports to generate personalised reports. Engineered a 4-tab analytics suite (Overview, Taste DNA, Listening DNA, Genre Evolution) with a radar-chart music fingerprint, personality typing across 5 behavioural axes, and genre evolution timeline. Deployed live on Streamlit Cloud.
0
2
Engineered a Closed-Loop Self-Validating LLM (CSVL) pipeline with a Generate→Critique→Refine architecture grounded in pre-computed statistical constraints. Designed a five-category hallucination taxonomy achieving∼10× hallucination reduction vs. single-pass baseline across 3 UCI datasets. Confidence scoring: conf = 0.6(1−p) + 0.4|r|. Accepted in IEEE GICITE 2026.
0
6
Deployed a content-based filtering recommender across 17,000+ anime, K-drama, and music titles using TF-IDF and cosine similarity. Engineered a 4-layer fuzzy matching pipeline achieving 100% dataset coverage and 95.6% genre overlap rate. Live on Streamlit Cloud.
0
12