GitHub Get in touch

Mert Cobanov

Senior AI engineer building agents with complex memory: backend systems, vector databases, diffusion model training, and generative pipelines. Mostly Python, FastAPI, and Pydantic AI.

Read the writing Get in touch

Writing about AI engineering, homelabs, and whatever I'm tinkering with

Writing

How I trained an object detector without drawing a single box Building a water bottle detector from scraped images and pseudo-labels, and what the three bugs along the way turned out to teach. TR 2026.08.23
How fast can a 27B model run at home? Running Qwen3.8-27B on four different machines, and what the numbers turned out to actually measure. TR 2026.08.19
KV Cache & Flash Attention LLM inference optimizations, one at a time: a smarter cache, a faster kernel, a better scheduler. 2026.05.19
How AI agent memory works Language models forget the moment they finish replying. Memory is everything the system around them does to make that not matter. 2026.05.07
From latent spaces to JWTs: how agents taught me backend I wrote code every day for years and somehow felt like I wasn't really writing code. Let me explain. 2026.03.28
See all 10 posts

Projects

Libraries

semaclust 5
pypi

Semantic text clustering using sentence embeddings and agglomerative clustering.
bucketsampler 2
pypi

Aspect ratio bucketing for diffusion model training. PyTorch native, DDP correct.
tasnif 35
pypi

Unsupervised image clustering with modern deep embeddings, PCA and K-Means.
depth2normal 115
pypi

Convert depth map images to RGB normal maps for shading and 3D workflows.
websum 124
pypi

Summarize web pages and YouTube videos with pluggable LLM backends.
kamera 5
pypi

Realtime webcam streaming and video processing with a FastAPI UI.

Contact

[email protected]

Based in Los Angeles. Open to interesting projects, collaborations, and long emails about AI, art, and tooling.

GitHub X LinkedIn