How my engagements work, and what you take with you.
How I structure engagements, what you take with you when we’re done, and why most operators have me run it.
Ownership
The work I build is yours.
When the build is done, you own the system. Not a login.
Every workflow I build is yours: the n8n workflow files, the Supabase database, the prompts, the documentation. Portable, exportable, version-controlled. If you want to take the system in-house, I support the smooth transition to your owned infrastructure.
Engagement shapes
Three shapes. Sequential by design.
Each tier does different work. Most engagements move through them in order. Full deliverables and scope live on the Services page.
Tier 01 · AI Foundation
Figure out what to build.
Four to six weeks. AI policy, tool stack rationalization, role-specific training, and a prioritized opportunity map.
Tier 02 · AI Implementation
Build it.
One to three months. Design, build, deploy, and stabilize the workflows. Strategic advisory throughout.
Tier 03 · Ongoing Partnership
Run it. Keep building.
Monthly retainer, no fixed end date. Operations, monitoring, iteration, new development, and Topo memory maintenance.
Start with the basics
Orientation on the present state of AI, and training on the fundamentals of using tools to support daily work.
For many teams, the starting point is education and training to get going. I lead half-day workshops for brands and teams that cover the current state of AI, how the tools work, how to use them, and specific training and demos tailored to your team’s work, including the systems they interact with daily.
A one-time, cost-effective way to start embracing AI tools, opening up bandwidth and building skills that pay off immediately and are vital for staying ahead of the learning curve as new capabilities roll out on an almost daily basis.
Operating model
How it runs.
The work is yours. The question is who runs it.
Default · Recommended for most operators
I run it for you on managed infrastructure.
Most teams I work with don’t have, and shouldn’t hire, a dedicated technical owner for an AI runtime. Workflows aren’t fire-and-forget. They run on infrastructure that needs maintenance and operational support. They drift as the business changes. They improve as new capabilities are retroactively applied.
The monthly retainer covers everything that keeps it running. You see the output. You don’t manage the infrastructure.
Alternative · The minority case
Your team runs it.
Available when your technical team owns and operates the runtime stack. I build workflows that deploy into your infrastructure.
Without that, workflows drift, the AI stack ages, security updates are missed, API call nodes break.
Retainer scope
What’s included when I run it.
The monthly retainer is the running gear plus the continuous improvement. Not a maintenance contract.
Operations and uptime
- Infrastructure operated on managed runtimes (n8n, Supabase, LLM APIs (Anthropic/Claude, OpenAI, Perplexity), PromptLayer, Better Stack, Topo — Loop Trails in-house runtime improvement engine).
- Heartbeat monitoring on every scheduled workflow with status page visibility.
- Backups and version control on every workflow and database change.
- Security updates and platform upgrades, tested in staging before production.
Resilience framework
- Uptime monitors with SLO tracking against the workflows that matter to your business.
- Self-diagnosing error detection: every failure is captured, classified, and routed to the right place automatically.
- Self-healing patterns where the system has solved the same class of error before — escalation only when something new lands.
- A real-time client-specific dashboard showing what workflows are running, their status, uptime, incident updates, and more.
Continuous improvement
- Monthly iteration on existing workflows based on usage and feedback.
- Topo memory maintenance: every approval and correction feeds the layer that makes month six sharper than month one.
- AI strategy advisory as the toolchain evolves. New models, new patterns, new integrations.
Pricing approach
How I quote.
Every engagement is priced against scope. I quote in proposals, not on a page.
True for everyone
- Tier 01 is a paid engagement. The fee is creditable toward Tier 02 if you continue within ninety days.
- Tier 02 is fixed-fee against the agreed scope. The stabilization window is included.
- Tier 03 is a monthly retainer. No fixed end date, sized to the workflows in production and the cadence of new development.
If you want a number to plan against, I’ll give you a range in the first conversation. The proposal that follows will be exact.
FAQ
Common questions.
Can I take everything in-house later?
Yes. The work is yours. If you ever bring it in-house, I help you transition: documentation, runbooks, access transfer. Some clients structure this from the start.
What happens if we stop working together?
Depends on which model you’re on. In managed, you get a full handoff: workflow files, database export, runbooks, access transfer. You bring it in-house or move to a new operator. In self-operate, nothing changes the day we stop. The system keeps running. What you lose either way is the iteration, the strategy, and Topo memory maintenance.
Do you work with our existing tools?
Usually. I integrate with what your team already uses. The build is workflow-first; the toolchain fits around the process, not the other way around.
Why is Tier 03 required after a Tier 02 build?
AI workflows aren’t fire-and-forget. They run on infrastructure that needs operations. They drift as the business evolves. They get sharper with corrections fed back into Topo. The Tier 03 retainer covers the running gear plus the continuous improvement. The alternative is self-operate, where Tier 03 is optional but workflows degrade without active stewardship. Either way, the build alone isn’t the end of the work.
What’s the smallest engagement?
A four-week Workflow Sprint inside Tier 02. One automation, scoped narrow, shipped end to end.
Start with a conversation.
Not a free diagnostic. A real conversation about whether the timing is right, what the highest-leverage workflow looks like, and how an engagement would actually run inside your operation.