Home › Running two businesses on one AI installation

Running two businesses on one AI installation

Affiliate disclosure: Mulkern AI Systems is an independent affiliate partner of Easy (easy.one), a product of Smart Apps Limited. If you buy Easy through our links, Mike Mulkern earns a commission, and the coupon applied at checkout (MICHAEL) is his. Everything on this page describes how we run our own businesses on Easy. It is our experience, not a statement by Easy.

Mike Mulkern runs two unrelated companies on one Easy installation: Mulkern AI Systems, which builds AI software, and Oil Rigs Now, an independent marketing platform for drilling rigs and oilfield equipment. One installation runs both businesses, with about forty scheduled jobs between them (counted 30 October 2026). This page is the map of how it is split, with links to each workflow.

Separate agents for each company

Each company has its own agents with their own roles. Oil Rigs Now has a lead Watcher, a Deal Gatekeeper, a market monitor, a social media drafter, and an operations supervisor. Mulkern AI Systems has its own web, video, product, and marketing agents. One coordinating agent sits above both and writes the daily report.

Keeping roles narrow makes each agent easier to brief and easier to check. When something goes wrong, there is one agent to look at.

Separate memory, one installation

Each business keeps its own knowledge in its own memory space, so an oilfield agent is not searching software notes and the reverse. Agents also keep their own working notes.

Easy describes one searchable memory across agents with shared spaces for things like a knowledge base or a client dossier, and projects that keep chats, agents, files, and knowledge separate. Separate memory spaces per company are the part we rely on most.

The workflows that run

These are the workflows we have written up, each from what actually runs on our installation:

• An AI chief of staff that writes the daily report for two companies
• A daily AI app factory that stops and waits for approval
• AI lead capture from email and web forms into a CRM
• A weekly AI SEO page pipeline that stages pages for approval
• An AI video production agent that checks every claim first
• Scheduled AI agents with approval gates: the rule we run under
• Running AI agents on flat-rate AI subscriptions
• AI social media drafts that wait for a person
• AI agent scheduled tasks: what 36 recurring jobs look like across two businesses
• An AI SEO audit agent that covers one property each week

What we would do differently

Start with fewer agents and write each role tighter. Several of our early agents overlapped, and two once claimed the same job. We fixed it by writing down who owns which lane.

Keep long conversations short. A chat re-sends its whole history every turn, so we start new chats for new topics and keep what matters in memory instead.

In the Easy documentation, see Projects and Workspaces and memory spaces.

Questions

More from how we run on Easy

• An AI chief of staff that writes the daily report for two companies
• Scheduled AI agents with approval gates: the rule we run under
• AI lead capture from email and web forms into a CRM