Our Work
Work That Ships
We build custom AI systems for leaders who need operational intelligence and agentic execution — not another dashboard. Every project here is built, deployed, and running in production. No mockups. No "coming soon."
Luna
AI-Powered Project Intelligence
What it does: Luna is a web-based control plane for multi-stakeholder projects. It synthesizes status, runs intelligent document search, and adapts task prioritization in real time.
Built for: Teams across time zones, disciplines, and organizational boundaries who need shared operational awareness without constant meetings.
Key features:
- Real-time project status synthesis.
- Natural-language document search across project artifacts.
- Adaptive task prioritization driven by dependencies and team capacity.
- Integrations with Asana, Google Drive, Slack, and custom data sources.
- Task lifecycle management: queue → active → signoff → archive.
- Crew dispatch with per-agent workspaces and context.
- Automated morning and evening briefing digests.
- Full audit dashboard: token usage, API costs, activity logs.
Meridian
Local Crew Cockpit
What it is: A native macOS desktop runtime for crew work. Meridian is where the captain dispatches specialists with slash commands, watches their output stream in real time, and intervenes when the situation calls for it. Built on Tauri v2 — Rust core, system WebKit, ~5MB binary — so it stays light, fast, and out of the way.
Built for: Operators running multi-project, multi-repo crew sessions who need a local cockpit for deep work, paired with a cloud control surface (Luna) for ambient command from phone, voice, or a brief laptop session anywhere.
Why it matters: Most AI tooling assumes a single chat-with-the-bot frame. Crew work needs a runtime: paired threads, streaming output, structured dispatch, offline-tolerant queueing, and a clean reconnect story. Meridian is that runtime — and Luna is its complementary remote cockpit.
LunarLegacies.com
Digital Legacy Platform
What it is: A Mālama-built product for preserving family stories, archival documents, and lived history as digital legacies that can be handed down across generations.
Built for: Families and individuals who want their record kept with care — not stranded on a dying drive or a forgotten cloud account.
Sidebar
Legal Case Management System
What it is: A case management tool for family law practitioners — a kanban-style case board with structured intake, document handling, and operational continuity built in. Currently running for an attorney managing 25+ active cases across five case types, built alongside her with weekly iterations and real user feedback.
Built for: Solo and small-firm attorneys who need software that respects how the practice actually runs.
- Kanban board with full case lifecycle management
- Auto-generated deadline sets per case type (8–13 deadlines per case)
- Document checklists with per-item status tracking
- Task system with assignees, priorities, and due dates
- Case notes and timeline with timestamped entries
- Master command center with smart alerts and overdue flags
- Google SSO with per-user data isolation
- Mobile-ready, works on phone in the courthouse
Quorum
Commitment-Driven Event Platform
Events only happen when enough people publicly commit. Organizers set a threshold; the platform handles the rest — tracking commitments, detecting quorum, and notifying participants automatically. Built for organizations that coordinate real action, not just RSVPs.
- Batch commitment UX — select multiple time slots at once (Doodle-style)
- Guest participation without forced account creation
- Automatic quorum detection with email notifications via Resend
- Cost tier support for events with per-person pricing
- Organizer dashboard with real-time commitment tracking
- Full event history and participant management
Kindling
Algorithmic Trading Engine
Paper trading engine for testing algorithmic strategies against live market data. Multi-strategy signal generation, real-time data streaming via Durable Objects, and position management with configurable exit rules. Demonstrates Cloudflare's Durable Objects for stateful, low-latency computation at the edge.
- Multi-strategy signal generation: RSI, MACD, Bollinger Bands
- Real-time market data streaming via Durable Objects
- Position tracking with stop-loss and take-profit rules
- Performance analytics and trade history
The AI-Literate Operations Layer — Kristin Kirk
Executive Coach & Leadership Consultant
Client: Kristin Kirk (Executive Coach & Leadership Consultant)
The problem: Kristin is a high-level strategist with deep expertise in leadership development. Like many visionaries, she was burning hours on AI tool friction — prompting, switching contexts, managing APIs, and duct-taping systems together.
The solution: We became her technical co-builder — the invisible AI-literate layer between her vision and execution — across three lanes:
- Content leverage — automated batch SEO optimization and podcast intelligence extraction, turning existing content into discoverable, high-value assets.
- Operational glue — custom agents sync her vision layer (Asana) with her execution layer (Google Sheets, CRM, email) so nothing falls through the cracks.
- Agentic critique — Sol audits marketing funnels, sales copy, and positioning in real time, providing strategic feedback without adding to her workload.
The result: Kristin focuses on coaching, content, and client relationships. The AI layer handles research, synthesis, distribution, operational continuity, and QA — a capable, invisible operations partner.
What she gets:
- More time in her zone of genius (coaching, not tool management).
- Operational continuity (nothing depends on her remembering to check a dashboard).
- Strategic intelligence on demand without hiring a full ops team.
How We Work
Discovery → Design → Build → Maintain
Discovery & Scoping
Week 1–2. Deep listening. What's the actual problem? What have you tried? What does success look like?
Design & Prototyping
Week 2–4. A working prototype fast — something you can touch, test, and give feedback on. No six-month waterfall.
Build & Deploy
Week 4–8. Ship iteratively. You get value early; we refine on real-world use.
Maintenance & Evolution
Ongoing. Systems evolve. We offer retainer-based support to keep your AI infrastructure healthy, updated, and aligned with where you're headed next.
Ready to Start?
Tell us what you're building or what's broken. We'll tell you honestly whether we can help.
Schedule a Discovery Call