AI engineer & researcher

Eray Yapağcı.

I build with language models.

My work spans AI agents, speech recognition, document intelligence, and making models run faster. Here are a few things I’ve researched and built.

Bilkent CS ’25 · Research & engineering at Wharton, BILSEN, and DataBoss.

01 / Publications

Papers.

ASE 2025

Agents in the Sandbox: End-to-End Crash Bug Reproduction for Minecraft

E. Yapağcı, Y. A. S. Öztürk, E. Tüzün · Accepted to IEEE/ACM ASE 2025 · arXiv preprint ↗

IEEE TASLP 2026

ADASTT: Adaptive Speech-to-Text Algorithm Selection via Meta-Learning

A. S. Namlı, O. F. Celik, E. Yapağcı, S. S. Kozat · IEEE Transactions on Audio, Speech and Language Processing, Vol. 34 · DOI ↗

02 / Selected work

A few things I’ve built.

Speech recognition · Open models

balbal-v2 →

Open Turkish speech recognition that runs on a CPU. balbal-v2 transcribes 236× faster than real time on 8 CPU threads; balbal-v2-manas makes the fewest meaning-changing errors of any open model on Mu-Bench. Weights on Hugging Face ↗, code on GitHub ↗.

Speech recognition · Moroccan Darija

balbal-v2e-zellige →

Experimental open speech recognition for Moroccan Darija, on a CPU. Lower error rates than Cohere Transcribe, Jais-ASR and MoulSot on Casablanca, MGB-5 and GigaSpeechBench MAR. Try it in the browser.

Open source

EasyOCR Optimized ↗

An optimized OCR pipeline with up to 3× faster CRAFT inference on large images.

Local inference

DeepSeek on Strix Halo ↗

A custom llama.cpp build for running DeepSeek V4 Flash on AMD Strix Halo, with optimized Vulkan kernels and speculative decoding.

03 / Get in touch

Say hello.

Want to talk about AI or something you’re building? Send me a note.