astrid greene

computer science @ university of michigan

astrid greene

computer science at the university of michigan, focused on software engineering and machine learning.

3.7
gpa
5+
projects
4
roles

// playable sudoku — try it

0/45 filled00:00
1
4
8
9
6
7
3
4
1
2
9
5
7
1
2
6
5
7
3
8
6
9
5
7
9
1
4
6
2
3
7
8
5
1
2
4
click a cell to start

About

background and focus

i study computer science at the university of michigan, focusing on data structures, algorithms, and systems. i build software across c++ and python, from low-level implementations to full-stack applications.

schoolgpahonors
university of michigan3.73/4.0university honors

currently focused on:

software engineering in c++ and python · core cs: data structures, algorithms, systems · designing efficient and usable applications

Experience

roles and impact

  • built an automated expense validation system in python across 5 azure function endpoints with ocr extraction, a 19-rule policy engine, and 10+ ai agent modules, processing 5k+ expense lines annually and saving 500+ hours and $35k+/year
  • built a deterministic gate routing system that examines report-level line composition and triggers only relevant ai agents per report, reducing unnecessary model calls and keeping agent analysis scoped to flagged subsets
  • designed deterministic-then-probabilistic validation architecture with worst-wins result aggregation, per-trip segmentation, and confidence-scored ocr fallback handling across 12 edge cases and 2 processing layers
  • wrote 500+ automated tests across 3 tiers using independent reference oracles and boundary-value coverage, deploying via github actions ci/cd with oidc auth, federated identity credentials, and azure key vault secret rotation
  • built a fastapi backend with sqlalchemy async, alembic migrations, and apscheduler running salesforce sync, deadline alerts (45/15-day), status reminders, and weekly pipeline summaries on configurable utc schedules
  • implemented 10+ specialized ai agent modules (hotel folio detection, duplicate receipt analysis, per diem reasonableness, etc.) with a strict safety invariant enforced in the parser: agent verdicts coerced to warnings
  • designed and implemented a role-based authentication system using supabase auth and postgresql, writing 10+ row level security policies to enforce granular access control across 3 user roles (admin, member, recruit)
  • developed a full-stack internal platform in react, typescript, and vite, featuring member management, attendance tracking, and event scheduling, serving 50+ active club members across 6 project teams
  • configured full-stack deployment pipeline using github, supabase, and vercel, managing environment variables and api keys across both development and production environments to support continuous deployment workflows
  • built an ai policy chatbot by migrating 200+ client policy documents into a structured google drive and connecting u-m maizey's rest api to embed a searchable chatbot widget directly on the client's password-protected site
  • resolved data ingestion issues caused by inconsistent document formatting, access-restricted pages, and duplicate file versions, cleaning and standardizing source data across 15+ file types to improve retrieval accuracy and reliability
  • conducted 6+ stakeholder meetings over a 4-week sprint to define system requirements, identify access constraints, and scope chatbot functionality, reducing the initial feature set by 40% to prioritize high-value policy retrieval
  • co-authored paper "debunking the curse of dimensionality in a k-nearest neighbors classification problem" with advisor dr. eric sakk, selected as a national semi-finalist in the junior science and humanities symposium
  • researched curse of dimensionality in k-nearest neighbors, running controlled python experiments to show that in uniform, hard-confidence data sets, increasing dimensionality can improve k-nn classification performance
  • designed k-nn experiments in python on datasets of 1000+ points, varying k-values and dimensions (2d to 15d) to test classification accuracy, with numpy, scikit-learn, and matplotlib for data generation, training, and visualization
  • led a 4-person team to design and develop a website addressing workplace discrimination
  • acted as lead debugger across front-end and back-end to support teammates and improve reliability
  • implemented a user story submission feature that collected over 100 contributions and published select narratives to spotlight underrepresented experiences

Projects

selected technical work

[featured] mar 2026

deep learning dog breed classification

custom pytorch cnn and vision transformer architectures for 10-class dog breed image classification across a 9,000-image dataset.

  • implemented transformer encoders, multi-head self-attention, and scaled dot-product attention
  • built transfer learning pipelines with adam optimization, checkpoint serialization, tensor normalization, and 256-patch embeddings, training models over 10,000+ update iterations

pytorch cnns vision transformers python

march 2026

personal website

personal portfolio built with modern web tooling. responsive layout, scroll and interaction-driven animation, and structured sections for experience, projects, and contact.

  • responsive design across breakpoints
  • animation-heavy ui with framer motion
  • structured content sections with clear hierarchy

next.js typescript tailwind css framer motion

github

feb 2026

icu mortality prediction model

clinical ml pipeline predicting icu mortality from sparse ehr time-series data across 12,000+ admissions and 40+ physiological variables.

  • transformed sparse ehr time-series data through normalization, imputation, statistical aggregation, and feature engineering workflows
  • executed 5-fold cross-validation and 1,000+ bootstrap resampling iterations across logistic regression, kernel ridge regression, and rbf kernel models, benchmarking auroc, sensitivity, and specificity under imbalance conditions

python scikit-learn numpy pandas

nov 2025

sudoku solver

c++ solver for 9x9 sudoku. represents the board as a 9x9 grid, validates rows, columns, and 3x3 boxes, tracks candidate values for open cells, and solves via repeated constraint checks and possibility elimination.

  • 81-cell board with row, column, and box validity checks
  • candidate tracking and constraint propagation
  • solves nyt easy puzzles in under a second; runtime measured with chrono

c++ stl chrono

github

oct 2025

order book simulator

price-time-priority central limit order book matching engine in c++ using priority queue (heap-backed) bid/ask books for low-latency trade execution.

  • processed 1m+ orders and executed 760k+ trades in under 10 seconds (74k+ orders/s) with o(log n) average matching complexity
  • implemented limit and market order types with real-time bid-ask spread tracking, partial fill logic, and order cancellation, validating engine accuracy against 500k+ expected trade outputs with automated test scripts

c++ heaps data structures

june 2025

bst-based map container

binary search tree with sorting invariants and an ordered map adt for efficient key-value storage.

  • sorting invariants, traversal logic, functors, templates, and recursion
  • o(log n) insertion and lookup performance

c++ bst templates

may 2025

naive bayes text classifier

multivariate bernoulli naive bayes classifier to classify posts by topic using log-probability scores.

  • determined labels using highest log-probability score
  • abstract data types for efficient file parsing and word-frequency detection
  • generates per-label log-likelihood maps used for automated topic classification

c++ probability adts

oct 2021

tic-tac-toe

interactive tic-tac-toe game in python using turtle graphics. keyboard-controlled for both players.

  • board, xs, and os drawn with turtle; win and tie detection
  • scoreboard, reset, and replay with basic error checking

python turtle

github

Education

university of michigan

b.s. in computer science, minor in french

ann arbor, mi · expected graduation: may 2028

gpa: 3.73/4.0 · university honors


relevant coursework

operating systems · data structures and algorithms · machine learning · linear algebra · discrete mathematics · computer organization · object oriented programming · calculus i-ii · computer pragmatics · programming concepts

Skills

languages, tools, and focus areas

languages

python
c/c++

bash

sql

r

java

c#

javascript/typescript

html/css

swift

kotlin

go

developer tools

git
linux

pytorch

scikit-learn

numpy

pandas

matplotlib

postgresql

github

rest apis

node.js

react

azure

docker

vercel

vite

interests

avid sudoku solver
card game enthusiast

dedicated tennis player

music lover & passionate musician

Campus Involvement

girls in electrical engineering and computer science
tech+ consulting and development
eecs 201 instructional aide
math exam proctor

Contact

open to internships, research opportunities, and collaborative projects.

GitHubhttps://github.com/astridigreene
LinkedInhttps://linkedin.com/in/astridgreene
Emailastridig@umich.edu