Projects

01
Agent Orchestration
Healthcare AI · Operations
Orchestration Design Instruction Architecture Governance Systems

20-AGENT
OPERATING
SYSTEM

Designed and deployed a full multi-agent operating system for a healthcare AI startup's operational intelligence layer. 20 agents across 5 domains — Intelligence Synthesis, Pipeline GTM, Data Operations, Quality Assurance, and Infrastructure — running in coordinated production with no external orchestration tooling.

Core architecture: a T1/T2 dual-tier instruction system that separates guard rails and evaluation routing (T1, loaded in agent settings) from full workflow procedures (T2, loaded on demand). T1 ensures consistent behavior and inter-agent handoff compliance at near-zero token cost. T2 provides full operational depth only when the task requires it.

Built a governance reconciliation matrix across all agents, formalizing every inter-agent handoff schema, escalation protocol, and human-gated control point. Signal routing pipeline delivers research intelligence from synthesis agents to decision surfaces in sales and leadership.

Agents 20 active — 13 Notion-governed + additional CLI shadow agents
Architecture T1/T2 dual-tier instruction system · zero external orchestration tooling
Governance Notion-native · reconciliation matrix · all high-consequence writes human-gated
Domains Intelligence · Pipeline GTM · Data Ops · Quality Assurance · Infrastructure
Engagement Fractional AI Systems Design · ongoing
20
Coordinated agents across 5 domains
2-tier
Instruction architecture — guard rails separated from workflow depth
Zero
Uncontrolled writes — all high-consequence actions human-gated
02
Agentic Infrastructure
Internal · Agentic Infrastructure
MCP Server Build Claude API Integration Agentic Identity

MCP SERVERS,
IDENTITY
+ CLASSIFIER

Three interlocking agentic infrastructure projects built from scratch. Together they form a working proof of the "You Are an API" thesis: extract intelligence locked in black-box systems and give users ownership of the layer.

Document intelligence system — a Claude-powered classifier. A 4-layer progressive cascade processes documents from filename only (L0) through PDF metadata (L1), document structure (L2), and targeted content envelope (L3). Privacy-first: L0 is the default, content reading requires explicit user permission. Validated against 10,836 real documents. Deployed on Fly.io with FastAPI, SQLite (WAL mode), and adapter-based ingestion from Google Drive and local filesystem.

Agentic identity provider — SSO for AI agents. A portable identity document (PID) that agents authenticate against instead of each building its own siloed model of the user. Scoped per-agent authorization controls what each agent knows and what it can do. Exposes a native MCP server interface so any MCP-compatible tool can request identity context. Built on WebAuthn passkeys with audit logging, scoped access, and revocation.

Portable AI memory server — normalizes conversation exports from Claude, ChatGPT, and Gemini into a unified corpus accessible as typed MCP resources. Deterministic SHA-256 IDs enable cross-platform deduplication.

Stack Python 3.9+ · FastAPI · SQLite (WAL) · Fly.io · mcp>=1.0
MCP Servers 2 built — agent identity interface · portable memory corpus server
Claude API 4-layer progressive classifier · confidence thresholds · heuristic fallback
Scale 10,836 real documents validated · 20+ document types · 7 entity types
Schema 13-table SQLite — entities, documents, lifecycle rules, flows, shares, knowledge
Auth WebAuthn passkeys · scoped delegation engine · per-agent field-level access controls
Importers Claude · ChatGPT · Gemini — cross-platform memory portability
10,836
Real documents validated through classifier
2
MCP servers built and deployed
4-layer
Progressive classify cascade — privacy-first by default
03
AI Systems Evaluation
Internal · AI Systems Evaluation
Prompting Architecture Systems Evaluation Team Training

AI PROMPTING
PERFORMANCE
ANALYSIS

A meta-analysis of AI prompting behavior across 18 Claude sessions (March 2026), scored across 8 dimensions: efficiency, delegation quality, pattern recognition, scope setting, instruction precision, context management, escalation judgment, and output specificity.

Built as a live interactive dashboard — rolling average timeline, radar chart, per-dimension breakdown — and converted into a Notion teaching module for the client team. The analysis itself demonstrates the pattern: extract the intelligence locked in operational behavior, surface it as structured infrastructure.

Score arc from 6.5 → 8.5 over 18 sessions documents measurable improvement in delegation quality and scope setting as the primary leverage points. Used to inform prompting standards across all active agent fleets.

Sessions 18 Claude sessions across 5 project types
Dimensions 8 — efficiency, delegation, pattern recognition, scope, precision, context, escalation, specificity
Score arc 6.5 → 8.5 over 18 days · primary gains in delegation quality and scope setting
Output Chart.js interactive dashboard · Notion teaching module · prompting standards doc
Application Informs instruction architecture across all 20+ active agents in production
18
Sessions scored across 5 project types
8
Evaluation dimensions
+30%
Score improvement (6.5 → 8.5) over 18 days
04
Operational Platform
Entertainment · Education
Platform Build Cloudflare Workers 0-to-1 Launch

FULL OPS
PLATFORM
LAUNCH

End-to-end operations platform for a dance education company — built from concept to deployed in weeks, no framework, no build step. Six distinct modules covering the full student and instructor lifecycle.

Audio delivery is handled by a Cloudflare Worker backed by R2 object storage — protected routes, streaming delivery, zero server management. Onboarding flows use decision-tree logic to route new students to the right starting point. The choreography builder generates structured session plans from parameterized inputs.

Stack HTML/CSS/JS · Cloudflare Workers · R2 object storage
Modules Onboarding · Choreography Builder · Student Progress · Knowledge Drills · Outreach Tree · Audio Delivery
Deployed sbc.wherewithal.studio · Cloudflare Pages + Workers
Build time Concept to deployed in weeks · no external dependencies
6
Modules covering full student lifecycle
Workers
+ R2 for protected audio streaming
Weeks
Concept to production deploy
05
Knowledge Infrastructure
Healthcare AI · Internal Tooling
Internal Tooling Team Enablement Knowledge Systems

TEAM
KNOWLEDGE
SYSTEM

Knowledge readiness system for a healthcare AI company's go-to-market team — 158 questions across 9 categories covering GTM, technical thesis, objection handling, competitor analysis, and agent system operations. Built to ensure every team member can speak the product fluently in any conversation.

Includes track filtering (Core / GTM / Ops), expert mode (tighter timing, fewer hints), a rotating knowledge drop that surfaces new content each session, and a score history system. Deployable in minutes, no backend required.

Questions 158 across 9 categories
Tracks GTM · Core · Ops — with expert mode and score history
Stack Vanilla HTML/CSS/JS · Cloudflare Pages · zero backend
Purpose Team readiness · investor prep · objection handling · agent ops fluency
158
Questions across GTM, Core, and Ops tracks
9
Categories including agent system operations
Zero
Backend — fully static, instant deploy

READY TO BUILD
SOMETHING THAT
actually runs?

We design and build agent systems that operate in production — not decks.

Start the conversation →