# Katya Teodorovich

> SWE @ Airtable

Location: United States, United States
Profile: https://flows.cv/katyateodorovich

## Work Experience
### Software Engineer @ Airtable
Jan 2025 – Present | San Francisco, California, United States

### Undergraduate Teaching Assistant @ Purdue Computer Science
Jan 2024 – Jan 2025
UTA for CS 381 Analysis of Algorithms

### ML Software Intern @ DeGirum Corp.
Jan 2024 – Jan 2024 | Santa Clara, California, United States
• Optimized real-time object detection on RTSP Raspberry Pi camera feed to run at >30 FPS
• Doubled YOLOv8 pose detection speed by refactoring C++ detection result postprocessing
• Wrote platform-agnostic benchmarking scripts for inference on Coral Edge TPU, Intel NPU, and CPU

### Tutor @ Horizons Student Support Services -- Purdue University
Jan 2023 – Jan 2024 | West Lafayette, Indiana, United States
Tutored students in Data Structures & Algorithms, Python, Calculus I & II, and Linear Algebra.

### ML Software Intern @ DeGirum Corp.
Jan 2023 – Jan 2023 | Santa Clara, California, United States
• Improved quantized YOLOv8 mAP by 7% through adjusting the model architecture to minimize concat layers
• Integrated a C++ postprocessor for YOLO detection into official DeGirum PySDK, running at <1ms per frame to parallelize with on-device inference with edge AI accelerators
• Trained models for 5 detection tasks, and exported to 160 total model variants 
        -- Datasets: COCO, License Plate, Face, Hand, and Car detection
        -- Formats: yolov5nu/v5su/v8n/v8s, with SiLu/ReLu6 activations, to ONNX/TFLite/N2X
• Hosted a live seminar on YOLO quantization

### Undergraduate Data Science Researcher @ The Data Mine - Purdue University
Jan 2022 – Jan 2022 | West Lafayette, Indiana, United States
• Collaborated with Battelle
• Developed scaleable NLP hyperparameter tuning script using HyperBand algorithm
• Balanced runtime and performance of Random forest, K-means, and Hierarchical Clustering tuning algorithms

### ML Software Intern @ DeGirum Corp.
Jan 2022 – Jan 2022 | Santa Clara, California, United States
• Automated label cleaning script for issues in training datasets such as Google’s OpenImages
• Removed double labels and flagged incorrect labels by comparing ground truth to inference results
• Optimized model performance by repeatedly training YOLOv5 model on custom data, then using inference results to clean original dataset
• Used K-fold cross validation to avoid model overfitting issues on small datasets


## Education
### Bachelor of Science - BS in Computer Science, Mathematics
Purdue University

### High School Diploma
Fremont High School


## Contact & Social
- LinkedIn: https://linkedin.com/in/kteodorovich
- GitHub: https://github.com/kteodorovich

---
Source: https://flows.cv/katyateodorovich
JSON Resume: https://flows.cv/katyateodorovich/resume.json
Last updated: 2026-04-11