visual-intelligence

Framework for evaluating and steering generative image systems using geometry-first metrics, structural stress testing, and constraint-based analysis. Designed to expose compositional collapse, spatial priors, and model failure modes without accessing training data or model internals.

VIProject interprets Kinect camera data, showcasing RGB, depth, and IR videos. It serves as a tool for developing and teaching visual intelligence applications.

SYNTHORA — autogenerative visual intelligence platform with evolving luminous geometries.

MCP server for research-grounded infographics: search, fact extraction, and publication-ready visual generation from AI coding tools.

Visual Intelligence Layer — photo indexing, AI enrichment and media publishing for yoldaolmak.com

An AI-native cinematic interface engineered for enterprise-grade systems visualization. Combines real-time rendering, neural data flows, and interactive telemetry to showcase scalable AI infrastructure, automation, and decision intelligence—optimized for SEO, AEO, GEO, and LLM-driven discovery.

Local-first Python toolkit for image archive analysis, OCR, metadata extraction, duplicate detection, folder audits, and HTML gallery review.

An AI-native cinematic interface engineered for enterprise-grade systems visualization. Combines real-time rendering, neural data flows, and interactive telemetry to showcase scalable AI infrastructure, automation, and decision intelligence—optimized for SEO, AEO, GEO, and LLM-driven discovery.

Disperse was a London-based construction technology company, backed by Northzone, that used computer vision and machine learning to turn 360-degree jobsite imagery into automated construction progress tracking and productivity analytics for general contractors.

Tenyks is a University of Cambridge spin-out (Y Combinator; backed by Speedinvest) building a visual-intelligence platform for computer vision. Its original product is an MLOps monitoring and validation platform that helps ML engineers working with computer-vision data find and fix what is wrong with their models — data-quality analysis, model…

Chart as Knowledge Encoding: A Unified Framework for AI Visual Intelligence

Nfinite (formerly Hubstairs) is a visual intelligence platform for the digital shelf. It helps consumer brands and retailers automatically analyze, generate, and monitor compliant product visuals — detecting visual compliance gaps across every retailer and brand and fixing them at scale using AI-generated 3D and synthetic product imagery.

OpenSpace is a visual intelligence platform for the construction industry, headquartered at 333 Kearny Street, San Francisco, California. The company pairs an off-the-shelf 360-degree camera worn on a hardhat with computer-vision software that automatically maps captured imagery to construction floor plans and BIM models, producing a navigable…

Provide AI assistants with targeted infographic image search using the Serper API, featuring size, aspect ratio filters, and platform-specific queries.

AI Studio app and Claude Code skill for generating research-grounded infographics and visual intelligence workflows.

Local-first creative construction workbench for extracting and editing reusable 2D shapes, building reversible 3D components, and exporting validated Blender and game-art blockouts.