I run a portfolio of products.

Style Guide

Building a design system from scratch takes weeks of picking colors, pairing fonts, and testing combinations. Style Guide turns a plain-language description of your brand into complete design systems — colors, typography, shadows, and spacing — ready to use in minutes. Refine through conversation, branch in a visual tree, and export to any platform.

• •Describe your brand, get complete systems — Describe your product in a few sentences and get full design systems back, with color psychology and WCAG contrast ratios already applied.
• •Refine by talking, not tweaking — Say "make it warmer" or "more contrast" in plain language. Every change updates 45+ components live, no hex values to edit.
• •Branch without losing work — Every variation is a node in a visual tree. Try new directions from any point and jump back in one click.
• •Preview on real components — See your system on buttons, cards, inputs, alerts, and more. Know exactly how it looks before you ship.

WealthPlan

WealthPlan is a Canadian mortgage vs. investment simulator for homeowners. Plug in your mortgage, your extra cash, and your province. See whether accelerating your mortgage, loading up your TFSA, or maxing your RRSP leaves you with the most wealth — with 2025 tax brackets, TFSA and RRSP rules, and year-by-year projections.

• •Answer the question every homeowner asks — Should I pay off the mortgage faster or invest the difference? Get a personalized answer grounded in your province, income, and mortgage.
• •Built for Canadians — 2025 federal and provincial tax brackets. TFSA and RRSP contribution rules. Refund reinvestment logic. Accurate where generic calculators guess.
• •Four strategies compared side-by-side — Mortgage acceleration plus TFSA, mortgage plus RRSP, pure TFSA, pure RRSP. See how each plays out over the life of your mortgage.
• •Year-by-year wealth visualization — A four-line chart shows how each strategy grows your wealth over time, plus a yearly breakdown table for the detail-oriented.

monday2github

This product catalog could not be loaded at build time. It will appear once the product endpoint is reachable.

C3

C3 turns your always-on dev machine into an autonomous agent platform. Slack alerts and Discord messages trigger headless Claude Code sessions that investigate errors, trace code, create PRs, and report back. Watch the agent work live from your browser, or continue the conversation from your phone. Open source (Apache 2.0), bring your own machine, your own tunnel, your own channels.

• •Runs on YOUR machine — Not a sandboxed cloud agent. It has your repos, your secrets, your CLI tools, your database access. Same permissions as you.
• •Channel-triggered sessions — Slack alerts and Discord messages start headless Claude Code sessions. The agent investigates and reports back in the same thread.
• •Watch it work, resume anytime — Every session is streamed to a web UI and fully resumable. Agent ran at 3am? Open it on your phone at 9am and keep going.
• •Encode your expertise as prompts — Prompt templates turn your internal playbooks into agent instructions. A 200-line prompt tells the agent exactly how you investigate an outage.

HireScout

HireScout connects to your ATS (Ashby first, more coming), enriches each candidate with Apollo and GitHub, and uses RAG plus GPT-4.1 to score them against the criteria you care about. You describe the role in YAML, HireScout returns a ranked, filterable dashboard with strengths, concerns, and tailored interview questions for every applicant. Built for hiring managers who are drowning in hundreds of applications and need a faster way to find the signal.

• •Plugs into your ATS — Ashby first, with Greenhouse and Lever on the roadmap. Pulls candidates, applications, and resumes directly through the API so nothing has to be copied by hand.
• •Enriches every candidate — Apollo.io verifies contact info and company details. GitHub surfaces real code, languages, and contribution history. Enrichment runs in parallel and is cached with sensible TTLs.
• •Scores against your criteria — RAG over candidate context plus Azure OpenAI GPT-4.1 grades every applicant on the categories and weights you define. Swap criteria and re-rank without re-pulling data.
• •Configurable per job — Each role is a YAML file: required skills, preferred skills, seniority, location, scoring weights, custom prompt. Version it with your team and keep evaluations consistent.

Impact Pulse

A survey platform built for nonprofits to measure program outcomes. Collect participant feedback, track impact over time, and generate reports that funders actually want to see.

Digitizer

Drop in a PDF or image — printed, scanned, faxed, rotated, multilingual, even handwritten — get back clean structured JSON. A hybrid pipeline combines classical OCR (Tesseract OSD, bbox geometry) with a vision-language model (Qwen3-VL-235B) so each step uses the cheapest tool that solves it correctly. Every field comes back with a bounding box, so the UI can highlight exactly what the model saw. API-key authenticated and agent-ready out of the box.

• •Handles anything you point at it — Handwritten cursive, sideways scans, smudged receipts, faxes, multilingual invoices, free-form letters. No template, no schema, no labels required.
• •Two-tier rotation, never sideways — Tesseract OSD handles the 80% case for cents. A VLM fallback judges orientation when OSD's confidence drops. Pages always come back upright.
• •Tight bounding boxes via a second-pass locator — The VLM's bboxes drift 5-10%. A focused locator call snaps each box to the cell after extraction, so highlights line up with the source.
• •Structured JSON, not OCR text — Every field is named, typed, and tied to coordinates. Items become tables, lists become numbered lists, prose stays prose. Render it, ingest it, agent it.

OneOps

We are no longer writing code. We are summoning it. OneOps is the practice that makes the summoning reliable. Declare the company in code. Let AI flow into every space the structure reveals. Humans return to the work only humans can do.

• •Predictable first — Deterministic code as the default. Infrastructure, identity, secrets, source control all declared.
• •AI second, in fixed roles — AI deployed only where the specification declares it.
• •Agents for the outliers — Triggered by unpredictable events the deterministic system cannot handle.
• •Four method layers — Foundation, Delivery, Orchestration, Intelligence.