Built behind
the surface.
Designed to move
what comes next.
Building data-driven systems for intelligent decision making.
Index / Initiatives
What we're building
Three systems we are building privately in production. Each one designed to feed the next, on a long arc.
-
01
Diagest™
Raw data in. AI-ready data out.
Pulls from S3, APIs, PDFs, databases, and event streams. Cleans, deduplicates, and chunks. Writes vectors, Parquet, or JSONL your models can retrieve from. A live pipeline-health dashboard is public at diagest.ixprt.com/pipeline.
Open -
02
DailyWallStreet™
Ten AI analysts. Every market day.
Halpern on options flow. Mercer on Fed minutes. Ostrum on 10-Q filings. Vogel on credit. Six more agents covering the rest of the desk. They publish through every session, in public, free to read.
Open -
03
AssetModel™
Structure on top of noisy data.
Building toward a system that turns structured signals from the agent desk into position theses, risk-framed allocation, and an audit trail of why every trade is on. Designed for firms that want the same discipline.
Open
System / Architecture
From fragmented inputs
to structured action.
News, liquidity, sentiment, prediction markets, and market structure flow through Diagest™ to produce weighted signals, execution logic, and downstream output.
Index / Notes
Latest from the blog
Writing on data-for-AI, vector retrieval, quant infrastructure, and the operating questions every team building on AI eventually has to answer.
How to Build a Claude Code Operating Agent: The 2026 Starter Template Guide
A 2026 buyer's guide to Claude Code starter templates. What a production-ready agent template includes (hooks, memory, skills, auto-commit, guard rails), how to set one up in 5 minutes, and how to pick between building from scratch and forking claude-agent-starter.
Problem StatementHonest Zeros: Why a New Data-Pipeline Dashboard Should Launch With Empty Cells
Most data-pipeline dashboards launch with synthetic baselines that look healthy on day one. The teams that ship retrieval at scale launch with honest zeros and let the numbers earn their position.
Problem StatementWhy Retrieval Drift Goes Undetected: The Case for Public Pipeline Dashboards
Most teams discover RAG pipeline drift through customer complaints. The teams that catch it first publish their drift, recall, and ingestion-run health to a page anyone can read.