Aug 2025 – Present · 9 mos · Chicago, IL · Hybrid

Mar 2025 – May 2025 · 2 mos

• Course: MACS 30113 1 Principles of Computing 3: Big Data and High-Performance Computing for Social Scientists
• Assist the instructor during lectures by offering technical support and helping students troubleshoot code and understand key computational concepts.
• Offer weekly office hours to provide one-on-one help with course material, programming assignments, and questions related to big data tools and parallel computing.
• Lead and manage weekly lab sections for over 30 students, guiding them through hands-on exercises involving big data analysis, parallel computing, and high-performance environments.
• Evaluate assignments and final projects for a class of 30+ students, ensuring consistent grading and delivering constructive feedback to enhance learning.
• Support students in configuring and using cloud-based and university HPC environments (e.g., AWS, Slurm clusters), and resolve related technical issues.

Aug 2024 – Sep 2024 · 1 mo · Chicago, Illinois, United States

• 𝐂𝐨𝐮𝐫𝐬𝐞: MACS 30120 - Computing Fundamentals Boot Camp
• Hold regular office hours to assist students with Python programming concepts, data structures, and algorithms.
• Debug and troubleshoot students' code during lab sessions.
• Grade weekly programming assignments using established rubrics and provide constructive feedback on code quality, efficiency, and style.

Jun 2024 – Aug 2024 · 2 mos · Chicago, Illinois, United States

• Collaborated with a professor in coursework development to integrate advanced social network analysis techniques into the curriculum, facilitating hands-on learning experiences.
• Developed a comprehensive document on SNA methods tailored to social media and an R Markdown tutorial on identifying and quantifying ego networks and polarization, incorporating practical examples and code implementation.
• Designed a capstone project that involved exploring Twitter data, visualizing networks, and generating SNA measures, providing students with hands-on experience.

Aug 2024 – Jun 2025 · 10 mos · Chicago, Illinois, United States

• Led end-to-end analysis of Chicago 311 service request data to evaluate the impact of housing policies on landlord behavior, leveraging Python for large-scale data processing and modeling.
• Applied time-series and spatial analytics in R to identify neighborhood-level trends in housing quality, urban stratification, and policy outcomes.
• Designed and implemented a Retrieval-Augmented Generation (RAG) system by embedding property tax PDF documents into a vector database (QuadrantDB), enabling natural language querying via GPT-4o API.
• Built an interactive chatbot to query and summarize property tax records, translating complex, unstructured PDF data into actionable insights for researchers and policymakers.
• Engineered feature extraction pipelines from semi-structured PDF data (e.g., property attributes, assessments, temporal indicators) to support downstream analytics and intelligent retrieval.

Feb 2022 – Jul 2023 · 1 yr 5 mos · Nagpur, Maharashtra, India · On-site

• Built scalable ETL pipelines processing 1M+ records, improving data processing efficiency by 40% and enabling downstream ML workflows.
• Developed and deployed ML-driven NLP chatbot systems, including Transformer-based multilingual models, improving response quality and reducing translation errors by 25%.
• Designed and automated A/B testing frameworks to optimize recommendation strategies and support data-driven product decisions.
• Deployed ML pipelines on AWS (EC2, S3, SageMaker, Lambda) using CI/CD, reducing production downtime by 30%.

Aug 2021 – Dec 2021 · 4 mos · Chicago, Illinois, United States

• Conducted social media network analysis focusing on political polarization.
• Applied NetworkX and Gephi for network visualization and statistical analysis to compute and assess diverse network metrics such as degree distribution, closeness centrality, and betweenness centrality.

May 2021 – Dec 2021 · 7 mos · Richland, Washington, United States

• Developed graph-based deep learning (GCNN + LSTM) models to forecast curbside parking availability, achieving 90%+ prediction accuracy.
• Benchmarked advanced ML models (N-BEATS, LSTM) against classical approaches, improving performance by 10% over best baselines.
• Designed end-to-end pipelines for time-series prediction, anomaly detection, and mobility demand forecasting.

May 2020 – Jul 2020 · 2 mos · Amherst, Massachusetts, United States · Remote

• Utilized ’tidycensus’ package in R to extract socioeconomic and demographic data of two metro-cities (Boston, Phoenix) from the US Census datasets and performed EDA, plotting geographical heatmaps to find correlation with e-Bird checklists data.
• Executed extensive data preprocessing and wrangling procedures, coupled with in-depth visualization techniques and hypothesis formulation, leveraging datasets including IHDS and NCRB.
• Explored correlations between crime rates and factors such as unemployment rates, education levels, and women empowerment, providing valuable insights into socioeconomic dynamics.