Shivani

Hi! I'm Shivani

A UW–Madison student studying Computer Science + Statistics, interested in technology-driven social impact, whether that's through code, data, or research.

Open to Summer 2027 internships in software engineering and data.

Feel free to reach me at spotnuru[@]wisc.edu

experience

Where I've worked

  1. May 2026 → Present

    Data Operations Intern

    Madison, WI
    TruStage · Filene Research Institute
    • Built an engagement dashboard that puts Salesforce campaign data and GA4 in one view
    • Wrote Python reconciliation scripts that cut 5 to 6 hours of cleanup per cycle
    • Mapping data flows for the Salesforce to Data 360 migration
    • Piloting Tableau Next workflows with the analytics team
    SalesforceGA4PythonTableau
  2. Feb 2026 → Present

    Undergraduate Research Assistant

    Madison, WI
    MadAbility Lab · UW–Madison CDIS
    • Testing where vision-language models break down on audio description
    • Studying what blind and low vision users need from a description, not what the model defaults to
    • Analyzed 500+ Meta Quest apps against a structured MR accessibility framework
    • Contributing to a paper in preparation with the lab
    HCIAccessibilityLLM EvalMR/AR
  3. Jan 2025 → Present

    Vice President

    Madison, WI
    Design Interactive · UW–Madison
    • Lead a 14-member exec team across strategy, sponsorship, and finances
    • Run agile workflows that keep semester-long projects on schedule
    • Oversee 3+ Madison-area client projects each semester
    • Help teams carry user needs from discovery through delivery
    LeadershipProductUX Strategy
  4. Jun 2022 → Sep 2024

    Data Science Intern

    Middleton, WI
    LAVIS Research Informatics
    • Automated 10+ recurring workflows across 3 clinical studies in Python and R
    • Cut audit prep time by 45% while holding 100% IRB compliance
    • Wrote 25+ SQL validation checks that caught bad data before analytics saw it
    • Built 6+ Tableau dashboards for cross-study monitoring
    SQLPythonRTableauETL
  5. Jun 2022 → Feb 2024

    Research Intern

    Santa Cruz, CA
    Tech4Good Lab · UC Santa Cruz
    • Co-authored "Exploring Communal Gratitude in Online Communities," PACMHCI at CSCW 2025
    • Ran thematic analysis on user research data
    • Turned the themes into design implications for prosocial platform features
    HCI ResearchThematic Analysis

about

About me

Hi! I'm Shivani. I work on software, data, and HCI research. I like building things end to end, and figuring out who they're actually for while I do it.

I got into tech because of the digital divide, the gap between who a technology is built for and who can actually use it. That's still what decides what I work on.

My first real exposure to HCI was co-authoring a paper with the Tech4Good Lab at UC Santa Cruz. I ran the thematic analysis, and it changed what I wanted from the work. I wanted to build with a specific person in mind.

At the MadAbility Lab, I'm testing where vision-language models still fail blind and low vision users. I'm also getting more into computer vision and responsible AI.

I'm always looking to apply my skills to new projects to solidify my learning. Right now I'm working on Pharavo, a medication companion for ESL patients and older adults. Check out my projects to learn more.

I'm also a data operations intern at Filene Research Institute, a credit union think tank that turns research into new ideas credit unions can test and put into practice, and vice president of Design Interactive, a design org on campus that works with local community partners on their products.

Outside of all this, I like , , , , and .

The Wisconsin State Capitol on a sunny day
Design Interactive members at the Design Showcase
Me as a kid with my family inside the Wisconsin State Capitol
The Design Interactive executive board on a staircase

University of Wisconsin–Madison

B.S. Computer Science & Statistics

Sep 2024 – May 2028 · L&S Honors Program

projects

Things I've built

Feb 2026 – Present

Pharavo

An AI medication companion for ESL patients & older adults

Medication non-adherence is one of the most expensive, avoidable problems in healthcare — and it hits ESL patients and older adults hardest, where a label they can’t fully read becomes a daily point of failure.

React NativeGemini APIOCRNLP
View case study
Sep 2025 – Jan 2026

Wisconsin Conservation Voices

Redesigning the Native Vote & Voting Info experience

The Native Vote and Voting Info pages carried critical information but buried it under confusing navigation. The ask: make it obvious, trustworthy, and fast to act on.

UI/UXFigmaUsability
View case study
Jan 2025 – May 2025

Spectacle Health

An AI chatbot for navigating health insurance

Health-insurance documents are dense, inconsistent, and terrifying to search when you actually need an answer. Members give up before they find it.

RAGWeaviateHugging Face
View case study
Jan 2025 – May 2025

Billboard “One-Hit Wonder”

Modeling chart reappearance with logistic regression

Does a one-hit wonder ever come back? I wanted to know whether an artist’s single chart appearance could predict a future one — 50 years of Billboard data, tested.

RPythonSpotify APIScikit-learn
View case study

media

What I'm into lately

I like keeping a log of what I listen to and watch. Listening updates live. Films come from my Letterboxd diary.

Listening
Last played
Mystery of Love
Sufjan Stevens
Call Me By Your Name (OST)
Recently
  • Vienna
    Billy Joel
    2h ago
  • Pink + White
    Frank Ocean
    4h ago
  • Cherry Wine
    Hozier
    yesterday
Watching
Recently watchedLetterboxd →
  • The Housemaid
    2025
    ★★★½
  • Dhurandhar
    2025
    ★★★½
  • Eternal Sunshine of the Spotless Mind
    2004
    ★★★★½
  • Call Me by Your Name
    2017
    ★★★★
  • Beautiful Boy
    2018
    ★★★½
Shivani's listening
Top tracks · last 4 weeks · sample
  1. 01ViennaBilly Joel3:34
  2. 02Pink + WhiteFrank Ocean3:04
  3. 03HeavenNiall Horan3:18
  4. 04Cherry WineHozier4:01
  5. 05Lover, You Should've Come OverJeff Buckley6:43
Top artist
Sufjan Stevens
Top genre
indie folk
Total
20:40
thanks for listening

stack

Toolkit

Languages
General-purpose language. Daily driver for data, ML, and backend work.Structured Query Language. Pulling, joining, and validating data.Statistical programming language. My go-to for regression and data viz.Object-oriented language. Class projects and algorithms coursework.Low-level systems programming. Memory, pointers, the classics.The language of the web. Interactive UIs and scripting.Typed JavaScript. Catches bugs before they ship.The building blocks of every web page I make.
Technologies
Cross-platform mobile. What Pharavo is built on.Component-driven UI library for the web.React framework for fast, production-ready sites — including this one.Python web framework for building APIs quickly.Open-source vector database. Powered the RAG over insurance docs.Open-source model and embeddings hub.Python data analysis library. Dataframe-everything.Classical ML in Python — regression, classification, clustering.Container runtime. Ship the environment, not just the code.Utility-first CSS framework. Used to build this site.
Methods
Retrieval-Augmented Generation — grounding LLMs in your own documents.Natural Language Processing — making sense of text at scale.Scoring model outputs against structured criteria to find capability gaps.Extract, Transform, Load — the plumbing under every data pipeline.Business-rule checks that catch bad data before it spreads.Interviews and discovery work to learn what people actually need.Shipping two versions and letting the data pick the winner.Watching real people use it. How the WCV redesign hit a 91 SUS.
Tools
Open-source relational database. Boring — in the good way.Lightweight embedded database. Perfect for prototypes.Document-store NoSQL database.Cloud infrastructure. EC2, S3, Lambda, the usual suspects.JavaScript runtime. Server-side JS and build tooling.Version control. Branches, PRs, the works.Where every design starts. UI, prototyping, and handoff.BI dashboards. Built 6+ for cross-study clinical monitoring.Google Analytics 4. Web engagement data for the TruStage dashboard.CRM platform. Campaign and audience data at Filene.

let's build
something useful.

spotnuru[@]wisc.edu