I help teams use AI and automation
without losing control.

I build practical multi-agent and workflow systems (n8n-style, MCP, Grok/Claude/ChatGPT) with shared memory, second-model checks, and a human approval step before anything leaves. I’ve run this on my own company first, documented the failures, and can show others how to do the same safely.

01
Diagram of the n8n workflow. Step 1: request form, where someone asks a research question. Step 2: fetch the top 3 Wikipedia search results. Step 3: an AI model writes the draft. Step 4: the workflow waits for Chris’s approval in Gmail. If approved, the draft is marked ready to publish to Google Docs. If rejected, it is stopped and nothing is saved or sent.
Diagram of the real n8n workflow (from the exported file). Open full size

Research → Draft with approval gate

A workflow that takes a research question, pulls sources, drafts a summary, and stops for a human to approve or reject before anything is saved or sent.

What it shows: Practical automation + clear human gate.

02

Second model check

One agent drafts. A different model reviews it for factual gaps or over-claims and flags them. The human sees both.

What it shows: Verification instead of trusting a single AI.

Status In progress

03
Diagram: Grok, Claude and ChatGPT each connect over MCP to Holdfast, which holds shared notes and rules and also works with other MCP assistants. Underneath is Hermes / Watchtower, Chris’s local control layer for memory, background jobs and tools on his own computer.
Diagram of my own setup. Open full size

Shared memory across AI assistants

Holdfast, my MCP memory and rulebook layer, gives Grok, Claude, ChatGPT and other MCP assistants the same notes and rules. It runs through Hermes, my local control layer with Watchtower, which handles memory, background jobs and tools on my own computer.

What it shows: Continuity across AI tools without re-explaining context.

How I work with teams

I start with one real job that already wastes time. We map what must stay human, build the smallest useful automation, and leave the team with something they can run and understand. I prefer short hands-on sessions over long training decks.

Open to remote roles with good overlap to UK/European hours.

Email: [email protected]

What I can do

Core

  • Human-in-the-loop design

    Agents in my system can propose changes but cannot deploy, widen their own access, contact anyone, publish or spend.

  • Multi-agent role design

    I run my company’s AI staff as separate roles, each with one written handbook, one job type and a human sign-off before anything goes out.

  • Local control layer

    Hermes, run through Watchtower, keeps my memory, runs background jobs and uses tools on my own computer; Holdfast and my agents run through it. Built and working. The memory and background-job tests below ran on Hermes.

  • AI memory outside the chat provider

    A decision saved to Google Drive came back into a fresh session with its source attached. Tested.

  • Durable background workflows

    A background job finished while I kept working, and a fresh controller retrieved the result. Tested.

  • Output checking

    A guard discarded a small model’s uncertain answer and sent the work back for review rather than accepting it. Tested. A separate second-model reviewer is in progress.

  • n8n workflow building

    Research to draft with a human approval gate (form, sources, AI draft, Gmail approve/reject, branch). Built; diagram above is from the exported workflow file. Two more demos in progress.

  • MCP and multi-model setups

    Holdfast, my MCP memory and rulebook layer, gives Grok, Claude, ChatGPT and other MCP assistants the same notes and rules. Built and working.

Supporting

  • Testing and honest status

    32 application tests passed; every capability is labelled tested, partial, experimental or planned.

  • Evidence-led research

    Research keeps what a source says separate from what the data supports, with gaps left visible.

  • Cost control

    Bounded jobs ran on a local model at no API cost, escalating to a stronger model only when needed.

  • Private remote access

    Service restart and private HTTPS access to my own machine tested.

  • Explaining technical tools to non-technical people

    Royal Navy submarine IT and communications, then B2B account management.

  • Process diagrams

    Labelled diagrams of my own workflows, drawn from the real workflow files (above).

HERMES / BUILT & USED BY CHRIS

Your AI changes.
Your system stays.

I built a personal intelligence system behind my AI chats. It keeps my memory, runs background jobs and uses tools on my own computer.

An authorised AI is the way in. The memory and work stay with me.Proven through local Hermes / Remote Desktop. Direct hosted ChatGPT, Grok and Gemini connections are not yet fully verified.

01 / YOUR SYSTEMKeep the context.
Change the conversation.
Concept illustration · not a product screenshot
Independent memory. Background work. Human decisions.

SEE WHAT I BUILT / 59 SECONDS

A system that stays with me.

Memory, background work and the decisions behind Hermes.

Silent film · AI-generated illustrations, not a product recording. The build log below records what is tested and what remains partial.

01 / HOW I BUILT IT

Change the AI.
Keep the useful part.

My design choice was to separate the AI from the memory and the work. An authorised assistant can use the same tools; each new connection still needs its own test.

ILLUSTRATE A DIFFERENT AI

ChatGPT: tested through local Hermes / Remote Desktop. Direct hosted attachment is not yet verified.

YOUYour question. Your decision.
YOUR CHOSEN AIChatGPT
THE PART THAT STAYSYour personal system
MemoryJobsAlertsTools
Your computer + connected servicesPermissions stay with you.

Interactive explanation. Switching a label here does not connect an AI account.

02 / IT REMEMBERS

Change AI.
Keep your memory.

The decision you made. The correction that mattered. The project you put on hold. Relevant context returns with its source, even in a fresh conversation.

Memory lives outside the chat provider. Facts, your decisions and AI interpretations keep their own labels. Sensitive records stay out of ordinary context.

How continuity was tested ↗
ConversationDecisionCorrectionEvidence
YOUR CONTINUITYOne memory.
Many conversations.
Source attachedCorrections retained
Fresh chatRelevant context

A selected record, not a copy of every private conversation.

03 / IT WORKS WHILE YOU DON’T

Leave a job.
Carry on with your day.

Long work gets its own place in the queue. You keep talking. The job continues, the result is saved, and you can come back later.

YOUR CONVERSATIONIllustrative example

Analyse recurring QUEST sellers.

Job accepted. You can carry on.

What did we decide last time?

Your decision, with its original context.

A six-second illustration of the queue. QUEST analysis remains partial. No real job is submitted.

  1. 01AskA bounded piece of work
  2. 02Hand offDurable job accepted
  3. 03Keep talkingWork continues independently
  4. 04ReturnResult + memory + useful alert
THE RESULT FINDS ITS WAY BACKYour conversation stays open.Press “Watch the hand-off” to follow one piece of work.
TESTED, NOT JUST AN IDEA

A diagnostic job continued while another query returned, then a fresh controller retrieved the saved result.

Read the test ↗

Small jobs can use a local model. Bounded tests ran with $0 API cost. When an answer needed stronger reasoning, the guard returned it for review. The test and its limits ↗

04 / IT WATCHES WHAT MATTERS

Better questions.
Evidence before certainty.

Three real research streams from Chris’s system. Each separates what a source says, what the data supports, and what still needs an answer.

01 / RAVEN + QUEST

Analyse recurring QUEST sellers.

Public ecosystem sources, pool observations and recurring-seller research. The question stays open until the evidence is strong enough.

WORKING Collection and evidence records

PARTIAL History is incomplete. Seller exhaustion is not proven.

02 / ICP

Did ICP adoption become real usage?

A framework that follows announcements toward builds, pilots, deployment, users and economic activity. A headline is a starting point.

WORKING Official-source monitoring

PARTIAL Observed statements do not prove adoption or revenue.

03 / IMX

Does Immutable growth benefit IMX?

Separate ecosystem progress from token value capture. Follow fees, demand and the staking questions that could change the thesis.

WORKING Source collection and review framework

PARTIAL Token capture remains an evidence question.

Research examples, not investment recommendations. Diagram shapes are illustrative. This page has no live connection to private research or financial accounts.

FROM YOUR PHONEPick up where
you left off.
“How did that job go?”
SAVED RESULTReady when you return.
Private connection
YOUR COMPUTERFiles · tools · jobsStarts. Syncs. Catches up.

Conceptual flow. No private interface or host details shown.

05 / YOUR COMPUTER BECOMES THE WORKER

Ask from anywhere.
Keep the work at home.

The controller reaches Chris’s own machine through an authenticated private connection. Background work can continue after the conversation ends.

Service restart and catch-up have been tested. The computer needs power and a connection to run new jobs; a full cold boot is still to be checked.

TESTED Service restart and private HTTPS access.
PARTIAL Physical-phone access and notification receipt remain to be checked.

06 / IT KNOWS WHEN TO STAY QUIET

Lots of observations.
One reason to look.

A useful system shouldn’t become another noisy inbox. Alerts are collected in one place, repeated notices are grouped, and you can acknowledge what you’ve seen.

Failed work can ask for attention. Research keeps its uncertainty. Email and fitness alerts are future applications, not active integrations.

ObserveGroupPrioritise
REQUIRES YOUR ATTENTIONA background task needs review.

See the evidence. Decide the next step.

Illustrative alert. In-app alerts work; phone delivery still needs device acceptance.

07 / DREAMING · EXPERIMENTAL

Don’t just repeat.
Learn what to change.

The experimental reflection process reviews corrections, failures and repeated effort. It proposes lessons or improvements, with evidence behind them.

WorkResultCorrectionReflectionLesson
LEARN. PROPOSE. REVIEW.

Better next time.
With a person in the loop.

The system can suggest a change. It cannot grant itself more access, deploy consequential changes or turn its own inference into a fact.

EXPERIMENTAL / UNDER DEVELOPMENT

08 / ROOM TO GROW

A new need.
Another useful job.

Start with something specific. Add a bounded capability when it earns its place. Keep the same memory, permissions and record of what happened.

Research is the first working branch. Email, fitness, travel, files and everyday admin are possibilities to build and test in their own time.

THE SHARED FOUNDATIONPersonal intelligence
ResearchWORKING · PARTIAL DATA
EmailPLANNED
FitnessPLANNED
Travel + adminPLANNED
FilesPLANNED
BusinessADAPT TO A REAL NEED

09 / WHY I BUILT IT

I was tired of
starting again.

Every new chat meant explaining myself again. Useful work stayed trapped in old conversations. Long tasks made me wait, and keeping track of everything became another job.

So I designed a system with its own memory and a queue that keeps working after I leave the conversation. I connected local tools with cloud storage, then tested whether the context and results survived a fresh session.

I use it to keep decisions and follow research questions. Phone control is the next acceptance check. New needs can become new jobs, using the same memory and permissions. The useful parts can then be adapted for someone else.

The work, the tests and the limits ↗Chris Rothwell Founder, Acorn Labs

11 / WHAT THIS COULD DO FOR YOU

One useful part.
Built around your problem.

These are parts of my own system that might be useful elsewhere. Any adaptation would need its own scope and checks.

These are possible adaptations. Customer integrations, including inbox triage, still need to be scoped, built and tested.

TALK TO CHRIS

What keeps taking your
time or attention?

If something here looks useful for a problem you have, tell me about it.

Tell me about it ↗A conversation about what might be useful.