Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: multimodal
GCA Explained: A Task Contract That Fixes Spatial Reasoning in VLMs
The Geometrically-Constrained Agent splits task formalization from geometric execution, so VLMs reason about space better with no training. How it works.
6 min readHybrid Image Search at Scale: SigLIP, SPLADE, YOLOv8 and Gemini Captions
Design of a million-image search system that pairs dense and sparse embeddings with object detection, entity recognition, safe search and LLM-written descriptions.
6 min readSearching Video and Audio by Meaning: A Multimodal RAG Build With Gemini Embedding 2
Multimodal RAG with Gemini Embedding 2: one vector space for text, audio, video and PDFs, so no transcription step, plus chunking, ChromaDB and re-ranking.
22 min readAgentic Predictive Maintenance on OpenVINO: One Pipeline, Five Industrial Use Cases
How an OpenVINO and LangGraph pipeline turns predictions into evidence and tickets across pipelines, solar panels, power lines, audio and raster data on Intel hardware.
7 min readAI Video Needs Orchestration: Lessons From the VideoAgent Pipeline Design
AI video needs orchestration more than a better generator: global planning, coarse-to-fine retrieval, executable graphs, repair loops and VideoAgent's gaps.
8 min readInference Beyond Text: Tuning VLMs, Embeddings, Speech, Image and Video
Each modality has its own bottleneck. See how to tune VLMs, embedders, ASR, TTS, image and video generators with the right batching, kernels, caching and quantization.
11 min readWrapping Multimodal RAG in a Gradio Chat UI
Turn a CLI multimodal RAG engine into a Gradio chat app with drag-and-drop file upload, progress updates, citations and answers streamed as they generate.
8 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.