Tag
ai
21 posts · all tags
Five Frontier Models Walk Into a Forecasting Council
I stripped the heavy actor-simulation out of my geopolitical forecasting pipeline and ran five lineage-diverse LLMs as a council on Iran-Israel-US. The consensus was interesting. The disagreement was more interesting.
Inside the Simulation: How Geopol Forecaster’s Actors Think
A deep dive into the simulation architecture behind Geopol Forecaster — the 40+ actor personas, the sealed-off reasoning model, world state persistence, and the six analytical lenses that critique each other blind.
AI Forecast: Forced Regime Change in Iran at 4-9% Within a Year
The Geopol Forecaster pipeline assessed the probability of forced regime change in Iran — IRGC removal, non-autocratic replacement, and formal Israel recognition — at less than 1% within a week, 1-3% within a month, and 4-9% within a year.
AI Forecasting the Iran-Israel Ceasefire: A 4% Chance of Holding
Running the Geopol Forecaster pipeline on the April 2026 Iran-Israel-US ceasefire produced a stark verdict: 55% at 24 hours, 22% at 72 hours, and just 4% at one month.
Geopol Forecaster: An Open-Source AI Geopolitical Prediction Pipeline
Introducing an open-source two-stage geopolitical forecasting pipeline that combines actor simulation with multi-lens analytical deliberation — built on tools the intelligence community is already experimenting with.
My Claude Code Projects Index & Plugin Marketplace
A look at two open source projects I've been building: a curated index of 200+ Claude Code project templates and a plugin marketplace with ready-to-install extensions.
Ivory Parts Finder: Israeli Computer Parts Price Comparison
A tool for scraping Israeli computer parts retailer Ivory with LLM-powered enrichment for price comparison against US retail prices.
Hebrew Text-to-Speech: A Provider Comparison
A snapshot of Hebrew text-to-speech capabilities comparing voice quality across multiple TTS providers including voice cloning experiments.
Hebrew AI Models: A Comprehensive Catalog
A catalog of AI models for the Hebrew language — LLMs, speech-to-text, text-to-speech, NLP, embeddings, translation, and more.
Jewish and Torah Tech Projects: A Curated Collection
A comprehensive curated list of technological projects for Jewish religious life — AI tools, Hebrew calendar utilities, text study platforms, MCP servers, home automation, and more.
Claude Code Lawyer: AI Legal Research for Israeli Law
A template repository for using Claude Code as an AI-powered legal research assistant, specifically tailored for Israeli law with Hebrew support and local legal databases.
Red Alert Monitoring Stack: A Self-Hosted Microservices System For Rocket Alert Awareness
A Docker-based microservices stack for monitoring Israel's Home Front Command alerts with real-time map visualization, AI situation reports, push notifications, and home automation integration.
Miklat MCP: An AI-Powered Bomb Shelter Finder For Israel
An MCP server that lets AI agents help users find public bomb shelters in Israel. 9 tools covering search, proximity lookup, filtering, navigation links, and neighbourhood stats.
Geopol Forecaster: A Multi-Agent Geopolitical Forecasting Experiment
An experiment in multi-agent AI forecasting: six independent analytical lenses generate structured geopolitical situation reports and scenario forecasts from live intelligence data.
MCPs, Skills And Agents For Israeli Developers
A curated directory of MCP servers, AI agent skills, and autonomous agents built for Israeli data sources — from bank scraping to bomb shelters, Knesset data to pharmacy lookups.
Personalised AI: a proposal for proactive, interview-based context generation
An architecture proposal for generating personalised AI context through deliberate interview-driven workflows rather than passive data collection.
Bias and censorship evaluation tests for large language models
Test prompts and notes exploring cultural bias, censorship, and worldview differences across Western and Eastern large language models.
Vibe Code and Learn: a subagent model to avert AI feudalism
Planning notes for a multi-subagent system that turns AI code generation into structured learning opportunities, not just output.
On MCP and consolidation: why the current setup is painful and what might fix it
Notes and resources on the pain points of MCP adoption and why consolidation through gateways and connectors matters.
MCP gateway model: notes on a better architecture for MCP management
Planning notes and sketches for an MCP gateway architecture that aggregates servers into LAN and WAN gateways.
ChatGPT Confirms: Ireland's Settlements Bill Is A Crock Of Sh*t
Even AI sees through the absurdity: An analysis of Ireland's controversial Occupied Territories Bill