The work behind the words

Don’t take the pitch.
Inspect the work.

Start with the problem you need solved. See what I built, why it matters, and where the evidence stops.

Map out your project ↗

Explore the interactive project studio

7 projects to inspect

macOS development build

NexusTerm

Agent activity is only useful if partial writes and old transcripts do not mislead the dashboard.

Inspect the reliability work

What I built: A native terminal project with transcript-backed activity, complete-record ingestion, and recorded-time tracking.
Why it could matter to you: Make agent work inspectable while separating historical activity from current work.
What you can verify: The macOS Xcode build and transcript regression checks passed, including every byte boundary of a Unicode fixture, file replacement, and stale activity.

Scope & limitations

Not a release or live dashboard recording. Session matching, large-file performance, and Swift 6 package compilation still need work.

Interactive web project

AI Operations Lab

A queue of customer messages is not a plan of action.

Try the working project

What I built: Source-linked task extraction, editable response drafts, human review, and a downloadable work plan.
Why it could matter to you: Less context switching between reading a request and deciding what to do next.
What you can verify: Explore a labeled fictional example. The application validates source references and supports an optional visitor-funded model request.

Scope & limitations

The example is not a client dataset or a measured efficiency result. Nothing sends automatically.

Independent engineering project

Sentinel

An assistant needs boundaries around the actions it proposes.

Inspect the engineering

What I built: Path validation, approval checks, secret-pattern redaction, and hash-linked audit entries around a local-model coding workflow.
Why it could matter to you: Make permissions and review explicit instead of relying on the model to behave.
What you can verify: The case study separates implemented application checks from the stronger isolation a production deployment would need.

Scope & limitations

Not a security certification, an OS-level isolation guarantee, or complete protection from prompt injection.

Independent control plane

ai-os

Background work needs verification and a way to escalate failure.

Read the operating model

What I built: Task orchestration with budget controls, verification commands, heartbeat monitoring, and escalation.
Why it could matter to you: Make unattended work inspectable and give failures an owner.
What you can verify: A first-person architecture case study describes the operating model and the failure that motivated it.

Scope & limitations

Private operating history is not an independently audited uptime or client ROI claim.

iOS development build

First Cut

A talking-head video should not require a desktop editing session to caption and trim.

See the build and tests

What I built: Native transcript editing, reversible cuts, three output formats, caption styling, and a one-pass video compositor.
Why it could matter to you: Turn editing decisions into text operations while preserving the original footage.
What you can verify: Nine simulator tests passed, including a real synthetic-video export with audio, captions, and a checked five-second result.

Scope & limitations

Working title. Not on the App Store; real-device transcription, long-clip performance, and paid AI calls remain to validate.

Local development tool

YouTube Video Factory

Video production spreads the same project across scripts, takes, edits, and publishing tools.

Explore the workflow

What I built: A local production workspace connecting scripts, transcripts, edit recipes, FFmpeg rendering, and optional publishing integrations.
Why it could matter to you: Keep a repeatable production process attached to each channel and project.
What you can verify: Implementation reviewed for transcript-edit planning and rendering integration. The case study describes the workflow and verification gaps.

Scope & limitations

Not a hosted public demo. No audience growth, publishing reliability, or time-savings claim has been measured here.

App Store portfolio

Native apps, publicly inspectable

Shipping means finishing the details beyond a prototype screen.

Open the app collection

What I built: Independent apps spanning learning, daily practice, documents, and geography games.
Why it could matter to you: A body of public product work a buyer can inspect before a conversation.
What you can verify: The collection links directly to App Store listings, including Bible365, Wandr, thinkr, and Sign Any PDF.

Scope & limitations

An App Store listing demonstrates distribution, not revenue, retention, or customer satisfaction.

Actual software. Actual screens.

Past the pitch deck.

Inspect the interfaces and engineering decisions behind the work. These are captures of running development builds—not generated product mockups.

Operations Lab · Silent, accelerated screen capture of the real interface using fictional messages. Load the queue, trace a source, review a response and challenge validation. No model request or external action was performed. Try the complete workflow yourself ↗

Video Factory’s working script-planning interface with a fictional weekly-update project — Video Factory · Actual local workspace, seeded with a fictional example. The capture shows planning UI, not a completed AI-generated production or a customer result.

FirstCut iOS editor running in the simulator with synthetic color-bar QA footage — FirstCut · Working-title iOS development build. Real simulator captures; color bars are a synthetic export-test fixture. Not an App Store release.

FirstCut caption-style controls in the iOS simulator — FirstCut · Working-title iOS development build. Real simulator captures; color bars are a synthetic export-test fixture. Not an App Store release.

Use the live Operations Lab Scope your version ↗