The workspace where AI actually works.

TL;DR

• BYO everything: models, agents, infrastructure, keys. No provider lock-in.
• The core insight: the future isn't "AI features inside apps." The workspace itself becomes programmable.
• Existing AI tooling is developer-native. Calyx makes agentic systems usable for ambitious non-developers.
• Building toward a future where people create their own tools, workflows, and mini-apps instead of buying fragmented SaaS products.
• The SaaS layer is becoming fluid. Calyx aims to be the environment where this new ecosystem lives.
• More than an app: long-term vision includes companion apps, ambient computing, AR interfaces, voice capture, and shared contextual intelligence across devices.
• Currently teaching these workflows professionally and operating at the frontier of AI-native work practices through Dragonfly Thinking and independent consulting.

The Problem

• Coding agents (Claude Code, Codex, Gemini CLI, OpenCode, etc.) have crossed an inflection point in capability.
• But the interfaces around them are primitive and developer-centric.
• Most knowledge workers still operate across fragmented SaaS tools and disconnected chat windows.
• AI systems become dramatically more useful with full contextual access — projects, files, tasks, notes, workflows — yet today's tools isolate context instead of integrating it.
• The best users are already hacking together IDEs, Obsidian vaults, scripts, and agents to approximate an AI-native workflow.
• There is no true home for human–AI co-creation yet.

Why Calyx

Calyx is designed as an actual AI-native operating environment. Local-first, extensible, programmable, markdown-native.

Built on a VS Code fork, which means:

• Native filesystem access
• Real execution environments
• Direct integration with coding agents
• Custom interfaces and mini-apps
• Support for local and cloud agents

Users can connect their own infrastructure: VPS-hosted agents, OpenClaw, Hermes, Claude Code, Codex, custom agents.

The goal is augmentation, not replacing humans with black-box automation.

The Big Thesis

The next generation of software won't be fixed SaaS products.

People are increasingly able to:

• Generate software themselves
• Orchestrate agents directly
• Create bespoke workflows for niche needs
• Customise interfaces in real time

The value layer shifts from:

to:

Calyx is designed to be that environment.

The SaaS Apocalypse

A reckoning is coming for SaaS. Apps that charge $20/month for narrow functionality are about to get vibe-coded out of existence.

People are already building their own replacements. Need a simple CRM? Vibe it. Social media scheduler? Vibe it. Invoice tracker? Vibe it. The threshold for "worth paying for" just jumped from "saves me time" to "genuinely impossible for me to build myself" — and that bar is rising every month.

Calyx isn't a victim of this apocalypse — it's one of the horsemen.

Mini-Apps + Ecosystem

A core part of the thesis is that users will increasingly build internal tools, dashboards, automations, content systems, and workflow-specific interfaces — directly inside their workspace.

Examples: CRM, social scheduler, research pipeline, invoice tracker, document extraction systems, internal dashboards, AI reporting interfaces.

Not separate SaaS subscriptions — composable mini-apps living inside the same contextual environment.

Illustrative example: A friend runs his own construction company. I built him a mini-app with a visual interface for agentic document extraction. His team uses it to pull and tag information from architectural door schedules — detailed spec documents that list every door in a building project. That feeds into an agentic pipeline that processes the schedules and automatically generates quotes. A workflow that used to take hours now happens in minutes. No SaaS product does this. No off-the-shelf tool ever will — it's too niche. But with the right platform, you can build exactly what you need.

This becomes both the moat and the GTM strategy.

Why Me

• 10+ years across startups, product, strategy, and marketing.
• Deep experience building operational systems inside tools like Notion, Obsidian, Roam, and AI workflows.
• Currently working at Dragonfly Thinking on strategic intelligence and multi-agent systems.
• Professional experience teaching AI workflows and helping teams adopt AI-native practices.
• Historically early to major shifts in AI interaction patterns and tooling.
• Strong product/design intuition combined with technical fluency and systems thinking.
• Know the personal knowledge management space better than most — early adopter of Roam, Obsidian, Notion.

Looking for a technical cofounder with deep expertise in cloud infrastructure, agentic sandboxing, security, and scale.

Where This Goes

Business Model

Use of Funds

Salaries — The AI talent we need is in high demand. Hiring generalist-ish roles: technical ability, design sensibility, and community instincts. People who can build, ship, and explain.

AI tooling and subscriptions — API costs, infrastructure.

Marketing — Portfolio approach: AI-generated content plus influencer marketing — paying creators to produce content showing how to do X, Y, Z where Calyx plays a starring role.

Traction

No paying users yet — by design. Launching with something far more advanced rather than shipping an MVP and immediately being in a race with copycats. The goal is to come out of the gate with enough depth that catching up takes real time.

Strong early signals:

• Dragonfly Thinking pilot — Currently exploring using Calyx as the demo environment for Dragonfly's multi-agent system. Dragonfly is a strategic intelligence company whose clients include government agencies in the UAE and Australia. If this moves forward, it's enterprise validation from day one.
• Demand is clear — Similar tools are starting to appear to great fanfare (Granola, etc.), but none are as feature-complete or architecturally ambitious as what we're building.

On Moats

Someone else could build this. The technology isn't proprietary. What I'm betting on is taste and timing.

I have a knack for product vision — better than most — and I've repeatedly demonstrated an ability to see where things are heading before they get there.

The moat isn't a patent or a network effect (yet). It's being the person who builds the right thing at the right time, with the right sensibility. That's how Cursor won. That's how Notion won. That's how we win.

More Than a Platform

Calyx isn't just a single app. Building out an ecosystem of companion apps: meeting recording, voice-to-text, capture tools, and more.

The magic is that all of these apps share context with your knowledge base. A meeting recording app that knows your projects, your tasks, your notes — it can automatically file transcripts, extract action items, and update your workspace.

Each app in the ecosystem is more powerful because of shared context. And that shared context makes the whole system exponentially more useful than any of these tools in isolation.