N
Niraj Bangari
About
Detail
Davis, California, United States
Résumé
Jobs
verified_user
0% verified
Data ScientistUS Army Corps of EngineersAug 2024 - Current (2 years 3 months)Built and deployed probabilistic and machine learning models combining Bayesian networks and NLP analysis of qualitative risk narratives to predict project cost and schedule outcomes; improved predictive accuracy by 20% over expert benchmarks, enabling data-driven decision support. Implemented MS Access/SQL tools to extract funding data from source files, resolving data pipeline failure. Created script to track operation and maintenance budget information for every project category, and developed interactive dashboards using Power BI to visualize performance, improving visibility into budget execution. Built SARIMAX time series forecasting model predict monthly project obligations, improving budget allocation accuracy, improving budget allo- Statistics Research ConsultantUC DavisApr 2023 - Aug 2023 (5 months)Performed linear regression and correlation analysis for a randomized controlled pilot study assessing guided imagery meditation and weight outcomes, identifying meditation frequency and demographic variables as significant predictors of weight and waist circumference changes. Published in the Pacific Journal of Health "Guided Imagery Meditation as an Adjunct to Weight Management" https://scholarlycommons.pacific.edu/pjh/vol7/iss1/14/
- Intern, AI Cloud TeamTenstorrentJun 2022 - Sep 2022 (4 months)Benchmarked ML inference workloads on Tenstorrent accelerators versus industry-standard hardware, contributing to internal evaluation of throughput and efficiency trade-offs for accelerator positioning. Utilized Docker and Kubernetes to containerize and orchestrate ML model testing on cloud environments, improving reproducibility and scalability of model evaluation pipelines. Developed and tested NLP and computer vision models across multiple architectures, supporting comparative performance analysis across model types. Conducted comparative analysis of AWS and Google Compute Engine, informing cloud platform selection based on performance, scalability, and cost-efficiency.
Education
verified_user
0% verified
- M.S - Statistics: Data Science TrackUC DavisSep 2022 - Dec 2023 (1 year 4 months)
- B.S - Statistics: Statistical Data Science TrackUC DavisSep 2019 - Jun 2022 (2 years 10 months)
Projects (professional or personal)
verified_user
0% verified
- SSentiment Analysis: Predicting U.S. Domestic Airline TweetsJan 2022 - Dec 2022 (1 year)Achieved ~90% accuracy and precision, demonstrating scalable sentiment monitoring for customer experience analytics. Built sentiment analysis models using Python and TensorFlow, combining classical ML and deep learning approaches. Provided a clear framework for selecting sentiment models based on data characteristics, accuracy requirements, and deployment constraints.
- HHigh Dimensional Time Series AnalysisJan 2022 - Dec 2022 (1 year)Analyzed high-dimensional dependent time series arising in financial and macroeconomic-style data, where the number of variables grows proportionally with the number of observations. Utilized Random Matrix Theory and an expansion of the Marchenko-Pastur law to study multivariate time series behavior. Enabled separation of signal from noise in high-dimensional time series, improving robustness of downstream forecasting and risk estimations.