Skip to main content
Products

Products, tools, and operational surfaces

This is the map of the systems I build on: where each product fits, what it owns, and which surface I reach for when the work moves from planning to production.

Compare

A map of the products and internal surfaces I use most often.

Validate

Each product links to docs, playground, and writing so you can inspect the work instead of relying on summary copy.

Deliver

In-progress work is labeled explicitly — nothing here is aspirational.

Use this page

Start with the surface, then follow the evidence

Open the product first, then follow the docs, writing, or playground path that shows how the work behaves in practice.

Loom SuiteCore

Loom

Editor-first MCP tooling with registry-driven config sync, shared skills, and agent context workflows.

  • Single registry across editors and assistants
  • Shared skill and context workflows
  • Consistent MCP server rollout patterns
Inference PlatformCore

FlexInfer

Kubernetes-native inference routing and deployment tooling with schema-backed configuration and operational workflows.

  • Routing + backend orchestration with clear config contracts
  • Predictable rollout, rollback, and capacity tuning
  • Docs and playground workflows stay in sync
Healthcare IntegrationCore

fi-fhir

Source Profiles, pipeline mapping, and HL7v2 to FHIR workflows for production-grade healthcare integration.

  • Profile-driven ingestion across legacy healthcare formats
  • Pipeline stages for parse, transform, and validate
  • Operationally friendly mapping and troubleshooting flow
Writing Hub

Read the tradeoffs and postmortems

If you want to see how I think, start with the case studies and writing that explain the decisions behind these product surfaces.

Playground Hub

Open the working surface

When the product card is not enough, the playground is where I keep hands-on validation flows and docs-backed tooling.

How I divide the work

Boundary matrix

Loom Suite
MCP config sync + multi-editor agent workflows
Loom, Loom Core, Loom Zed integration
Docs →Open →
FlexInfer
Inference routing and deployment control
Schema-backed config, routing, operational controls
Docs →Open →
fi-fhir
HL7v2/FHIR pipeline integration
Source Profiles, parser phases, mapping workflows
Docs →Open →
3

Core surfaces I use most often: runtime, context control, and sensitive-data integration.

2

Boundary types I validate first: deployment and integration.

In progress

Unfinished enterprise controls stay visible here so you can tell what is shipping and what is still moving.

Architecture

Architecture snapshots

High-level diagrams for inference routing, public operations surfaces, and GPU GitOps automation.

Multi-model inference platform

Routing layer, providers, and platform boundaries.

Multi-model inference platform

FlexDeck public API

Sanitized read-only API surface for live operations data.

FlexDeck public API

GPU GitOps platform

Flux-driven cluster automation and rollout controls.

GPU GitOps platform