Chaturya Yarradoddi
ML Engineer | RAG Systems · LLM Fine-Tuning · Predictive Modeling | MS Data Science @ UMBC | Azure DP-100 | Power BI Analyst (PL-300)
- Role
- Data Analyst at University of Maryland Baltimore County
- Location
- Baltimore, MD, US
- LinkedIn followers
- 500 followers
About Chaturya Yarradoddi
MS Data Science candidate at UMBC (May 2026) with hands-on experience building production-grade ML systems, LLM pipelines, and data analytics solutions. I specialize in: RAG pipelines — built a fully offline system using BM25 + FAISS + cross-encoder reranking with local LLaMA inference (published NLP research paper) LLM fine-tuning — Code LLaMA 7B & Gemma 2B using SFT, LoRA, and QLoRA workflows ML Engineering — end-to-end scikit-learn and PyTorch pipelines with MLflow tracking, feature engineering, and model deployment Predictive modeling — XGBoost graduation risk classifier on 25,926+ student records (AUC = 0.846), deployed with Power BI dashboards for executive decision-making Data Analytics — longitudinal financial aid impact analysis on 28,960+ records using SQL-driven preprocessing, cohort segmentation, and equity-focused inferential modeling SQL & Database — complex querying, schema design, and data pipeline development across PostgreSQL and MySQL for institutional and healthcare datasets Certified Azure Data Scientist (DP-100) and Power BI Data Analyst (PL-300).
Experience
Data Analyst
University of Maryland Baltimore County
Oct 2024 — Present · US
Led institutional-scale analytics initiatives to support data-driven academic policy, financial aid optimization, and student success strategy across student records.• Designed and validated an end-to-end predictive modeling framework on 25,926 multi-year student records to estimate undergraduate graduation probability, integrating academic performance, financial aid allocation, and enrollment pathways, achieving AUC = 0.846 and enabling institutional leadership to identify high-impact intervention thresholds.• Applied advanced feature engineering and cohort-aware validation to address class imbalance, data leakage risk, and distribution shifts across pre/post-COVID populations, ensuring model generalization and analytical rigor at scale.• Conducted longitudinal financial aid impact analysis using Expected Family Contribution (EFC) segmentation and HEGIS-based STEM classification, identifying a 44.9% STEM completion rate among supported students versus 30.8% among non-supported cohorts and up to 3× higher persistence among supported high-need groups.• Translated statistical findings into actionable institutional insights through interactive Power BI dashboards and policy simulations, informing financial aid redistribution strategies, academic advising interventions, and equity-focused student success initiatives.
Education
University of Maryland Baltimore County
Master's degree, Data Modeling/Warehousing and Database Administration
Chennai Institute of Technology
Bachelor's of Engineering in Electronics and Communication Engineering, ECE
2019 — 2023
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.