Live model · Executable billing sandboxPrototype
Extends the The Agent Harness field note
Runtime Policy Lab ↗
Give a real AI billing agent a task. Compare its tool calls and account changes with and without server-enforced controls.
Root18D / Product Lab
From exploration to working products
Ideas made tangible. Products you can try.
Most prototypes are small, deployed AI agents grounded in the engineering already documented in a Root18D project record or field note — try them directly, no signup required. A few are independent, downloadable open-source tools instead; each page states clearly which kind it is.
FormatLive demos & real tools
DataSynthetic, bounded, or none
Archive12 prototypes
DisclosureEach page states AI-generated or real code
01Product Lab
Live demoPrototype
Extends P05 Sales Engineering
FinOps Anomaly Copilot ↗
An AI agent that explains Azure cost anomalies in plain language and recommends concrete next steps, extending the FinOps advisory work in the Sales Engineering case study.
Live demoPrototype
Extends P01 RAG Onboarding
RAG Eval-as-a-Service ↗
Hosted faithfulness, relevance, and context-recall scoring for any retrieval-augmented generation pipeline, extending the grounding work in the RAG Onboarding case study.
Free preview / waitlist openPrototype
Extends P03 BETA-MIND
Model-Training Micro-Courses ↗
A structured course on LoRA, QLoRA, distillation, and RLHF, built from the training-the-delta field note, with one free sample module you can try right now.
Live demoPrototype
Extends P06 Global Support Delivery
Support-Ops Triage Assistant ↗
AI-assisted queue routing with retrieval-grounded runbooks, extending the standardized operating model from the Global Support Delivery case study.
Live demoPrototype
Extends P04 Cloud Capacity
Serving Cost/Latency Simulator ↗
Estimate inference cost and latency for a model-plus-GPU combination before committing infrastructure, grounded in real Azure GPU pricing and published memory-bandwidth specs from the serving-the-token field note.
Live demoPrototype
Extends P04 Cloud Capacity
Confidential-AI Readiness Checker ↗
Scans a sample deployment configuration against the confidential-computing production checklist and returns a prioritized remediation plan, extending the confidential-computing-inference field note.
Live demoPrototype
Extends P02 GHCP Adoption Telemetry
Copilot Adoption Analytics ↗
Interprets GitHub Copilot adoption and acceptance-rate telemetry for a synthetic team and recommends concrete next steps, extending the GHCP Adoption Telemetry project.
Live demoPrototype
Extends P04 Cloud Capacity
Azure Capacity Forecaster ↗
Forecasts weeks-to-quota-exhaustion from a synthetic subscription's usage trend using a transparent linear-regression model, extending the Cloud Capacity project.
Live demo · Downloadable sourcePrototype
Extends the The Agent Harness field note
Agent Harness Starter Kit ↗
A real, tested, downloadable reference implementation of tool contracts, a step/cost budget, an approval gate, and structured tracing — the four guardrails from the the-agent-harness field note, with an AI agent narrating a real captured execution trace.
Live demoPrototype
Extends P03 BETA-MIND
BETA-MIND Research Instrument ↗
An illustrative prototype of the BETA-MIND reflection-instrument design: pick one of four human-agency dimensions, write a short reflection, and get a Socratic follow-up — never a score or a diagnosis.
Open source · Beta v0.3.0Prototype
An independent open-source project
agentop ↗
A terminal tool for monitoring and safely controlling local Ollama model deployments: live GPU and memory metrics, safe model lifecycle actions, and an optional proxy that captures the request history Ollama doesn't provide.