Namit Shrivastava

Namit Shrivastava

he/him

Graduate researcher in survey methodology, trustworthy AI, and computational social science.

Survey Methodology • Data Science • Responsible AI

Currently building

Designing rigorous measurement systems for AI-enabled research.

Graduate researcher with proven expertise in survey methodology, causal inference, geospatial epidemiology, and advanced statistical modeling. Published survey research in peer-reviewed venues and presented at AAPOR on transformer-based sentiment analysis methodologies. Research agenda focuses on developing trustworthy data integration frameworks that fuse survey methodology principles with generative AI, deep learning architectures, and privacy-preserving computational methods to advance automated data collection, quality assurance, and responsible measurement at scale.

View CV Publications Projects GitHub
129K+ Census tracts analyzed
1.1M+ Social posts processed
2.4M Images in ML workflows
91.6% Best model accuracy

Links

Research Focus

Turning complex data pipelines into reliable measurement systems.

My work centers on building trustworthy data integration frameworks that connect survey methodology with modern AI systems, so automated collection, classification, and inference stay interpretable and empirically grounded.

01 • Collect

Survey-aware acquisition and quality control

Designing multimode collection systems, response-monitoring logic, and missing-data adjustment plans so inference starts from defensible measurement foundations.

Survey design Missing data Quality assurance
02 • Model

Large-scale spatial and social inference

Connecting administrative, geospatial, and behavioral signals into modeling pipelines that stay readable across public-health, survey, and computational social-science settings.

Geospatial modeling Epidemiology Causal inference
03 • Govern

Responsible AI and privacy-preserving deployment

Embedding confidentiality, interpretability, and auditability into NLP and machine-learning systems so automation can scale without weakening measurement trust.

Transformer NLP Privacy Responsible AI
18 datasets automated through CKAN lifecycle tooling
129,572 census tracts integrated in public-data research
1.1M+ social posts modeled in transformer sentiment work

Work + Education Snapshot

A compact pipeline from technical foundations to research systems delivery.

This view compresses the education and work journey into a readable sequence so the homepage feels directional instead of list-driven.

01 • Quantitative foundation

Birla Institute of Technology and Science (BITS) Pilani

Built the base in engineering, statistics, and machine learning through a civil engineering degree with a data science minor.

Nov 2020 - Jul 2024 Pilani, RJ, India
02 • Applied engineering

Legistify Services Private Limited

Moved into production-oriented ML and platform work across OCR, trademark similarity systems, API testing, and scalable automation.

Machine Learning Engineer Jan 2024 - Jun 2024
03 • Graduate methods depth

University of Maryland, College Park

Deepened survey methodology, data collection, and causal inference while also teaching privacy and confidentiality through JPSM.

Master of Science, Survey & Data Science (Data Science Track) Teaching and Graduate Assistant
04 • Research systems delivery

Institute for Social Research, University of Michigan

Scaled into large public-data research and infrastructure, from geospatial epidemiology at census-tract scale to CKAN-driven repository automation.

May 2025 - Dec 2025 Social Data Science Center, University of Maryland
Open full CV

Research Interests

Survey Methodology Causal Inference Transformer NLP Geospatial Analysis Privacy-Preserving AI Statistical Modeling Deep Learning Sentiment Analysis

Explore

Research Publications Conference papers and scholarly work on sentiment analysis, responsible data systems, and applied AI. Build Projects Selected technical projects spanning machine learning, forecasting, analytics, and engineering systems. Ship GitHub Repository spotlight across research code, workflow automation, and starred builds with live project links. Present Talks Conference presentations and talks translating research into practical, communicable findings. Teach Teaching Courses and instructional work focused on privacy, methodology, and computational practice.

GitHub Spotlight

Starred repositories that best represent current work.

The GitHub showcase now refreshes from the GitHub API automatically and highlights the repositories that map most directly to active research, portfolio work, and reproducible code artifacts.

Open GitHub showcase All repositories Prompt library

Jupyter Notebook · namo507/Smart-Order-Routing-SOR

Smart-Order-Routing-SOR

Public repository in the GitHub profile.

1 stars
0 forks
283 repo KB
View repository

CV Snapshot

100%

Broadband completeness

Achieved full broadband data completeness across 129,572 U.S. census tracts through multi-source fusion.

35%

Search discoverability gain

Improved open-data repository search discoverability through hierarchical taxonomy redesign.

90%

Maintenance reduction

Reduced manual maintenance overhead with bulk metadata automation and schema-compliance workflows.

844

Blood donations mobilized

Helped organize a two-day blood donation drive with 60+ volunteers and 1,000+ donor records.

Education

Master of Science, Survey & Data Science (Data Science Track)

University of Maryland, College Park

Aug 2024 - May 2026 • GPA: 3.814/4.0 • Dean's Fellowship Award Winner AY 2025-26

Bachelor of Engineering (Honours), Civil Engineering (Minor: Data Science)

Birla Institute of Technology and Science (BITS) Pilani

Nov 2020 - Jul 2024 • GPA: 3.327/4.0

Highlights

  • Published survey research in Springer and presented at AAPOR on transformer-based sentiment analysis.
  • Built ML systems processing 2.4M images, 1.1M social posts, and 129,572 census tracts.
  • Improved data discoverability by 35% and reduced metadata maintenance work by 90%.
  • Delivered quantifiable outcomes across research, infrastructure, public health, and ML engineering.

Contact

Open to research collaboration, data science work, and thoughtful conversations.

Reach out via email, connect on LinkedIn, or browse code on GitHub.