Amin Fadaeinejad·Nov 4The Fridge That Watches the DoorHow we built an AI fridge assistant with one camera, one Gemini call per door cycle, and seven small agents that each do one job.
Amin Fadaeinejad·Oct 1222. RoPE Part 2: From 3D Video Grids to FLUX.2’s 4D Position IDsPart 2 of a series on Rotary Position Embedding. Part 1 covered 1D RoPE and 2D axial RoPE for Vision Transformers.A response icon1A response icon1
Amin Fadaeinejad·Oct 1021. Rotary Position Embedding from 1D to 2D: Derivation, Intuition, and PyTorch CodeThis is Part 1 of a two-part series on RoPE. It covers 1D RoPE, why it encodes relative position, and its 2D extension for Vision…A response icon1A response icon1
Amin Fadaeinejad·Aug 2020. How Transformers Work: Building One Module at a TimeConsider this sentence:A response icon1A response icon1
Amin Fadaeinejad·Aug 819. When to use SAM, and when not toSooner or later, every team that touches segmentation problems asks the same question. “We need segmentation. Can we just use SAM?”
Amin Fadaeinejad·Aug 318. Per-Scene Optimization vs. Feed-Forward Inference: Two Ways to Reconstruct a 3D SceneTwo families of methods currently dominate 3D reconstruction from images, and they are built on opposite assumptions.A response icon2A response icon2
Amin Fadaeinejad·Aug 117. Unlocking Fast LLM Inference: A Deep Dive into KV CachingLarge Language Models (LLMs) have reshaped the AI landscape, but moving them from training to production exposes a severe efficiency…A response icon2A response icon2
Amin Fadaeinejad·Jul 2816. V-JEPA 2 in Practice: Loading, Feature Extraction, and Fine-Tuning the EncoderIn my previous article, I described what a Joint Embedding Predictive Architecture is and why V-JEPA 2 [1] predicts in a latent space…A response icon1A response icon1
Amin Fadaeinejad·Jun 2815. Can a Language Model Hear? A Technical Tour of Audio & Music Understanding LLMsA hands-on comparison of five open audio-language models such as Qwen2-Audio, Audio Flamingo 3, Gemma 4, MOSS-Audio, and MU-LLaMA on the…
Amin Fadaeinejad·Jun 2714. Fine-Tuning a Vision-Language Model with LoRA and QLoRA: A Hands-On GuideA vision-language model (VLM) takes images and text as input and produces text as output. Modern open VLMs, Qwen2.5-VL, Llama 3.2 Vision…A response icon2A response icon2