Skip to content
View adamblumoff's full-sized avatar

Block or report adamblumoff

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
adamblumoff/README.md

Current Project:

CareBase helps caregivers manage everything in one place: appointments, bills, and notes, so they can focus on patients instead of logistics. Caregiving often involves information overload from emails, texts, portals, and calls. CareBase aims to cut through that noise and reduce caregiver burnout by centralizing key information.


Other Recent Projects

Built a Python CLI system for large-scale LLM debate runs with reproducible configurations, structured JSONL/CSV artifacts, and automated scoring and summarization. Designed and implemented a read-only analytics dashboard that ingests artifacts from private object storage, computes metrics server-side, and renders interactive performance and cost views. Hardened the system with production-grade reliability features, including usage and cost accounting, retry logic, and immutable run snapshots to support consistent high-volume batch execution.

Built a machine learning platform that predicts at-risk K–12 students before they fall behind, improving early intervention and academic outcomes. The system utilizes an explainable neural network model that achieved an AUC of 81.5% on the validation data. It integrates real-time dashboards and intervention tracking across platforms such as Canvas, PowerSchool, and Google Classroom, enabling teachers to identify struggling students and take prompt action. Developed with Python, FastAPI, and PostgreSQL, with over 125 automated tests ensuring reliability and performance.

Developed competitive Othello AI engines using search algorithms like Principled Variable Search and Q-learning. The alpha-beta engine achieved a 90% win rate against a fixed opponent after over 1,000 simulated games. This project focused on comparing a state-of-the-art game-playing algorithm to reinforcement learning techniques, and concluding which is better for Othello. I also co-authored the paper, which is in the repo.


Recent Technologies and Tools

Python, TypeScript, T3 stack (kinda, but you get the point), React Native, Expo, PostgreSQL, S3 (buckets in general), Railway, Codex


Contact

Location: St. Louis, Missouri

Website: www.blumey.dev

LinkedIn: www.linkedin.com/in/adam-blumoff

Resume (As of 12/30/2025)

Pinned Loading

  1. cosc-257_bird_dbms cosc-257_bird_dbms Public

    Forked from seojinjung/cosc-257_bird_dbms

    A DBMS for a project studying social dominance in mixed-species flocks of birds. Final project for COSC-257.

    SCSS

  2. OthelloEngine OthelloEngine Public

    COSC-241 Final Project

    Python

  3. student-success-prediction student-success-prediction Public

    TypeScript

  4. VRBasketball VRBasketball Public

    VR basketball game trying to nail the physics of basketball.

    C#

  5. ZombieShooter ZombieShooter Public

    ASP.NET

  6. carebase carebase Public

    TypeScript