Decomposition, QA, final judgment, and review gates.
Agent workflow
How I structured my AI agent workflow
Hermes is where I started treating AI less like one chat box and more like a workflow system: profiles for different roles, memory for durable context, skills for reusable procedures, and review gates for judgment-sensitive work.
Profiles
Profiles turn AI work into assigned roles
The most useful Hermes feature to explain publicly is profiles. They make the workflow legible: different profiles are configured for different kinds of work instead of forcing every task through the same model, prompt, and context.
Input
David / project need
Receives the task, chooses the lane, and keeps the work tied to project context.
Coordinator
Default profile
Receives the task, chooses the lane, and keeps the work tied to project context.
Design polish, layout critique, hierarchy, and accessibility review.
Lower-cost inventory building, source clustering, and research passes.
Implementation experiments, alternate hosted models, and local model serving tests.
| Profile | Role in the workflow | Why it matters |
|---|---|---|
orchestrator |
Planning, decomposition, review, QA | Separates judgment and review from raw implementation. |
design |
UI/UX critique and polish | Gives interface work a dedicated review lane. |
knowledge-worker |
Evidence gathering and source inventory | Uses cheaper models where first-pass organization is enough. |
omlx-test |
Local model experiment | Explores privacy, cost, latency, and local serving constraints. |
minimax-m3-coding |
Coding-oriented experiment | Tests whether lower-cost coding lanes are useful for implementation work. |
Memory + skills
Making AI work durable instead of disposable
The goal is continuity. I do not want every AI session to restart from scratch, and I do not want repeated procedures to live only in prompt history.
Durable project context
Persistent memory and Obsidian-backed notes preserve decisions, source evidence, project conventions, and review constraints across sessions. That turns AI from a one-off assistant into a continuity layer for project work.
Reusable procedures
Hermes skills act like workflow playbooks: project planning, debugging, browser automation, deployment, testing discipline, and design review. When a pattern works, it can be captured and reused instead of reinvented.
Model routing + review
The right lane depends on the job
One of the biggest lessons has been that the answer is not always “use the strongest model.” The right choice depends on cost, quality, privacy, risk, and whether the work needs reasoning at all.
Takeaway
What this taught me
Enterprise AI is workflow design. Models matter, but so do state, review, logging, access control, repeatable procedures, and knowing when not to use a model at all.