Founder identity across the portfolio

The question people ask

The most common question I get about Mustard Seed Group is some version of: *what actually ties this together?*

On the surface, the scepticism is reasonable. There is a B2B operating system. There is a fitness platform. There is a research lab working on autonomous agents. There is a services division building custom AI systems for clients. There is a consumer media and community ecosystem. From the outside, it looks like a collection of side projects with a holding company name bolted on top.

That reading misses the thing that actually holds it together. Not a market. Not a theme. Not a technology stack. A person, and more specifically, a particular kind of person with a particular orientation to the world.

This letter is about that. About what it means to maintain a coherent identity as a founder when you are building across multiple domains at once. About what you lose if you do not take that question seriously. And about why I think capability, not product category, not audience segment, not revenue model, is the actual connective tissue.

What fragmentation feels like from the inside

There is a specific discomfort that comes with building across domains. It does not feel like confusion about what to do next. It feels more like a subtle erosion of the right to claim expertise.

If you are building a fitness platform, there is a kind of person who builds fitness platforms. If you are building B2B SaaS, there is a kind of person who builds B2B SaaS. The categories come with recognisable identities, credible communities, legible career arcs. You know what success looks like. You know who your peers are.

When you are building across all of these simultaneously, you occupy none of those identities cleanly. You are not quite a consumer founder. Not quite an enterprise founder. Not quite a researcher. Not quite an operator. The temptation, and it is a genuine temptation, is to pick one and pretend the rest are hobbies.

I have not done that. Not because I am contrarian, but because I do not think the domains are as separate as the categories suggest. And because the thing I am actually building is not any single one of them.

Capability as the spine

Every product in the MSG portfolio is concerned with the same underlying thing: what can a person or team do that they could not do before?

Orbit is not a CRM with extra features. It is an attempt to collapse the cognitive overhead of commercial execution: to give a founder or a small team the operating surface they need to run a lead-to-launched-product workflow without losing track of context, relationships or momentum. The product is useful because capability is limited and attention is finite.

CheekyGains, the consumer fitness product inside the All Purpose ecosystem, operates on the same principle from a completely different angle. Physical performance is a capability. Consistency is a capability. The coaching relationship, and Naira, the AI performance coach inside CheekyGains, extends this into the product, is about transferring structured knowledge and accountability into a person's daily behaviour. Different audience, different domain, same underlying question.

Benediction Lab does not ship consumer products. It researches the infrastructure of capability at the machine level: how agents retain context across sessions, how memory systems allow tools to accumulate useful knowledge rather than starting from scratch each time, how autonomous systems can produce reliable output without constant human supervision. The lab exists because those questions have to be answered before the portfolio's ambitions can scale.

TUXX, the services and custom AI division, sits at the intersection of research and deployment. It builds things for real clients in live environments, which means it tests patterns in conditions that simulated research cannot replicate. Pattern Up, the sub-product under TUXX, extends that into a more structured offering.

None of these products could exist without the others being developed in parallel. Not because they share a codebase or a customer base, they do not, but because they share a thesis about what is worth building and why.

The role of the founder as connective tissue

There is a version of the holding company structure where the parent entity is purely financial: a vehicle for capital allocation and portfolio management. That is not what Mustard Seed Group is, at least not at this stage.

Right now, the connective tissue is not a corporate structure. It is a person. The coherence of the portfolio lives in the founder's ability to move between domains without losing the thread: to apply a research insight from Benediction Lab to an Orbit product decision, to bring a lesson from a TUXX client engagement back into how CheekyGains thinks about its coaching model, to notice patterns across commercial and consumer behaviour that would be invisible if you were only operating in one of them.

This is not the same as being a generalist in the shallow sense. A generalist diffuses energy across domains out of curiosity or restlessness. The founder-as-connective-tissue does something different: they compress knowledge from multiple domains into a single coherent thesis, and they use that thesis to make better decisions in each individual domain than a specialist working only within it could make.

The capability thesis is what makes this compression possible. Because every product is asking the same underlying question, what can this person do now that they could not do before, insights transfer. The research on memory systems is relevant to how Orbit should handle long-running client relationships. The lessons from coaching individuals in CheekyGains are relevant to how TUXX thinks about onboarding clients into complex AI workflows. The domains look separate at the product level. They are not separate at the level of the question being answered.

What this demands of you

Maintaining that coherence is not automatic. It requires something specific from a founder in this position, and I want to be honest about what that is.

It requires that you stay close enough to each domain to understand what is actually happening in it, not at the level of status updates and dashboards, but at the level of the actual work. If Benediction Lab is running experiments on agent memory, you need to understand what those experiments are revealing, not just whether the team is on schedule. If TUXX is working through a challenging client environment, you need to understand what the client is actually struggling with, not just whether the project is on budget. That kind of proximity is how the connective tissue stays alive.

It also requires that you resist the pull towards platform thinking too early. Platform thinking, the instinct to abstract everything into a unified system as quickly as possible, is appealing when you are operating across multiple domains, because it promises to make the coherence legible to investors, partners and the market. But premature abstraction kills the individual insights that make each product genuinely useful. The portfolio needs to stay concrete long enough for the patterns to reveal themselves honestly, before you start designing the architecture that connects them.

And it requires a particular relationship with your own identity. You cannot afford to let your sense of self become dependent on the success of any single product. When you are the connective tissue, the failure of one node does not invalidate the structure, but only if you have built your identity around the thesis rather than the individual bets. If Orbit has a hard quarter, that cannot become a founder crisis, because there is a whole portfolio of other questions being answered, and those answers inform whether and how Orbit recovers. The portfolio has to be able to carry the founder as much as the founder carries the portfolio.

One thesis, many expressions

September 2025 is a particular moment in this work. The portfolio is not finished. It is not close to finished. Each product is still early relative to where it needs to be. The research is ongoing. The commercial patterns are still being tested and refined. There are decisions ahead that will be genuinely difficult.

But the thesis is clearer than it has ever been, which makes everything else easier to hold.

Mustard Seed Group exists to build systems that increase human capability. That is the sentence. Everything else is an expression of it: a different domain, a different audience, a different product form, a different stage of development. Orbit expresses it through commercial execution. CheekyGains and Naira express it through physical performance and personal accountability. Benediction Lab expresses it through foundational research. TUXX expresses it through live deployment and client environments.

The founder's job is to keep the thesis honest: to make sure that every product decision, every resource allocation, every partnership is actually advancing the question that the whole structure was built to answer. That is not a management task. It is an identity task.

The portfolio only makes sense if the centre is clear. The centre is capability. And right now, the centre is held by a person who is still, very deliberately, building across all of it at once.