Nicolas Cava

A system attempt for startup studios

Nicolas Cava
Nicolas Cava

Nicolas Cava

6 minutes read
A system attempt for startup studios
Picture by frank mckenna

A startup studio journey

Intellectual property as a service

One of the greatest force of a startup studio is the powerful intellectual property (IP) that it owns, which unlock startup scalability potential, engineer and customer high-quality experiences, and massive business technological strike capacities. We give them the nuke button.

The IP enable us to create products, in fact startups, the quickest with the minimum effort possible. That's a important divergence from traditional venture capitalists or incubators. One of a startup studio core mission is to grow a powerful startup ecosystem.

Based on this fact, IP's power is to give many years of technology advances on startups in regards to their concurrence on the same scale. Because of this, IP needs to be reevaluated on a day-to-day basis, just to maintain its technological advance. Experiments and tweaks need to be more natural than eat and breathe.

Scalability challenges of the mission

Having such large startups network add complexities. They evolve independently of each other, at their rhythm, with products that differ for each business or market they are trying to penetrate.

Even if the IP gives them a great concurrent advantage from the start face of their concurrence, its complexity can quickly become impossible to manage in terms of technical debt.

Even more, we don't want to constraint startups by imposing them technological limits to be able to protect a certain interference level.

For best comprehension of the current ecosystem architecture, we will name:

  • Core IP: the startup studio's IP that each startup starts from, given the current snapshot.
  • Startup IP: current startup IP, strictly distinct from the startup studio's one (core IP).
  • Engineering AI: engineer's personal bot that can generate core IP snapshots and emit strategic suggestions on codebases.
Core IP snapshot flow diagram
Core IP snapshot flow diagram

Startup studio's engineers work to simplify and evolve the IP. As world-class engineers, their role is to experiment strategies and technologies at the edge of the industry capabilities.

However, startups are not synced with the startup studio's mothership, as their IP, which becomes outdated from the core IP as soon as it will be modified by startup teams. Startup IPs can as well take advance from the core IP with the cross-pollination strategy, that enables teams to inspire each other with technological and strategical moves.

The cross-pollination strategy gives enough autonomy and freedom to teams that they start inspiring each other and reuse the more efficient technological and methodological experiments across the whole company ecosystem.

And with that, the startup studio should benefit from technological successes coming from its startups.

We can totally assume that startups start with a base core IP snapshot then work on top of it without never be synchronize again with it. But, we will lose the amount of generated values from the core IP since that time, within the effort of the startup studio's engineers. Values, that can be a great asset for startups, when they can evolve exponentially their technology without a single resource involved.

Based on the fact that each startup evolves at their rhythm and asynchronously from the startup studio, it's very difficult to predict and plan focused strategies of IP migrations and updates, because of bad timing, backlog prioritization, or low resources, and so on.

We don't want that the startup studio's and startups engineers employ themselves to copy-paste codebases which give zero overall output value.

We want to automatize and barbarize the maximum amount of boring work so startups can focus on their business context only. We want to create and share intelligence through our startup ecosystem with powerful flows, channels, and love! Finally, we want to give capacities to startups which they can do chirurgical strikes into their corresponding markets.

Let's find a way to build a powerful startup and IP ecosystem that avoids zero value output!

An ecosystem of only-valuable outputs

How can we gradually build such an ecosystem

Being very disciplined with the core IP architecture is the first step of the strategy. It enables us to well identify the whole component landscape so that we can process to make improvements on each component at scale without side effects.

Owning such discipline would make it possible to efficiently segment the core IP that each startup can reuse only the needed part of it (less code, less technical debt).

Core IP segmentation proceeds into two categories: packages and services. Trying to boost the "packages" category is a good objective because they can be reused and versioned easily by using packages manager like NPM. Packages are libraries or frameworks like React or Redux, and services are websites, apps, BFFs, or downstream services, they put together multiples packages to form a working software entity.

For example, Yarn can tell you which package is outdated, but there are not such out-of-the-box things for services, because they are not hosted on a package manager.

We want to isolate architectures in generic and reusable components, capable to be downloaded each time with needing them to generate a new product. When the core IP is updated, the generator will download the last version each time, so new startups will have the best IP they can own within the startup studio's ecosystem.

Day-one startup IP setup with core IP components
Day-one startup IP setup with core IP components

How can we reach the mandatory discipline level to unlock this kind of architecture

The first step, conventions' step. Have strict conventions but sufficiently flexible to cover every possible development case is a big step.

We speak about programming language conventions, like JavaScript or CSS style guides, but architecture, infrastructure, implementation, conventions as well (and even more).

The next step, it's conventions enforcement between engineers. This passes through the best humanly possible way indeed, and by code reviews.

But, it will be in the first place automated, and engineering AIs should play a great game here. They can check convention rules depending on the context based on the core IP's one, emit warnings, suggestions, or block builds and deployments when necessary. To be able to do this, we need better data as well as good algorithms.

The final step is probably the more important: engineering culture.

If engineers share the same powerful culture, they can hold a high discipline level on core IP architecture. They can be able to find the best strategies to simplify this mission when creating strictly standalone services.

Culture is foundation here, this strategy cannot be done by a one-man army, team building and highly talented engineers that work very well together is mandatory given the discipline level involved.

What is going on when we want to update an outdated core IP snapshot into its last version

We can humanly analyze the core IP again, infer that changes need to be done, or communicate through Slack or email the new versions or migration plans, but that will not be well scalable within the kind of huge startup ecosystem we want to build (imagine thirty startups, with many products and teams by startup!).

So we should extend our favorite development tools, to add an intelligence layer which can compare the core and the startup IP together, and automatically suggests migration plans to corresponding teams. Like that, teams will be always aware that their IP needs to be updated and tools should tell them the best way to do it efficiently (automatically if they can, if not, manually).

Developer tools would no longer be simply tools but real AIs, personal engineering bots, like the Jarvis of the engineer!


It will take care of ungrateful tasks without or with less output value, and unlock humans to the possibilities to focus on their business.

Engineering AI (Jarvis!) will reveal that a startup IP is outdated
Engineering AI (Jarvis!) will reveal that a startup IP is outdated

In synthesis, we can see the core IP like the central intelligence and technology of the startup ecosystem, that each startup would draw from it to multiply their capacities.

Another level to explore to glorify this ecosystem

How can we motivate engineers to perform their best in this ecosystem? To be able to do that, we should work on the belonging notion, on its readability, its progression level, and short and long-term strategies. We can inspire ourselves with video game patterns.

Video game industry has, once again, broken psychological truths and patterns about humans, to the profit of experience enhancement and universe belonging. Gamification is the trendy term, which applies perfectly to our context.

In our industry, especially engineers are attracted by gamification concepts, because this speaks to their cognitive capacities, coming from sciences and mathematics.

With this example, each core IP's component is named, with statistics, objectives (missions or quests), and a strong identity, which multiply culture among engineers in order to succeed this strategy.

Even more, own data around the core IP unlock engineers to take better strategic decisions regarding the IP's future: what are the trends, weaknesses, strengths, etc.

Give a name on each component, give us an identity, missions will improve reading capacity and vision of the core IP, and will augment inspiration.

Inspire ourselves from gamification concepts will grant the core IP a soul, heat frozen and strict architecture notions, inspire culture, and enable engineers to see the whole ecosystem like an organic and living entity that they are part of.

Would this be sufficient to own the magnificent ecosystem we want?

Probably not, the ecosystem will probably evolve within our experience and experiments.

However, today, build this strategy with the core IP and engineering AIs bring important foundations that will revolutionize our way to make products and startups.

Finally, the success of this strategy establishment is based on core IP capacities and intelligence, and, on top, engineers power and culture in the startup's ecosystem.

Let's discuss building better companies.

You have better chances of finding me active on Twitter. I tweet about leadership, remote work, software engineering, personal growth, and systems. I also build my company publicly.

I'm also active on LinkedIn, where I post longer thoughts.

You can also contact me by email.

Recent posts

Recent publications on all platforms. Follow me on Twitter and LinkedIn.