The Tree Before the Company

Avocado is not a clever acronym or a market-tested brand name. It is the memory of a lived experience.

In the city where the founder grew up in the Amazon region of Brazil, there was an avocado tree. He climbed it. He picked the fruit himself rather than always waiting for some to do that for him.

When the moment came to name a company, he was finally willing to start, after years of hesitation, that childhood image of reaching for something himself was the one that stayed with him.

The founder describes a specific and candid reason for waiting as long as he did to become a founder.

He could see very few visible examples of Black founders who had built something at scale, not in his immediate surroundings and only rarely online.

That absence of reference points fed a quiet but persistent belief that access to investment, and perhaps success itself, was structurally harder for founders who looked like him.

For years, the response to that belief was deferral. Maybe later. Maybe later. Until a little over a year or two ago, he stopped deferring.

That delay, and the reasoning behind it, is not incidental colour rather, it is a direct data point about how perceived representation gaps shape founder behaviour long before any product, market, or investor enters the picture.

---

The Core Problem

Avocado AI exists inside one of the most crowded categories in applied AI right now, generative tools for marketing and creative content.

The founder is unambiguous about this. Competitors copy feature releases within days. The market is saturated with tools that can generate an image or a video.

The actual problem Avocado AI targets sit one layer beneath the obvious one. Agencies and e-commerce brands creating advertising content face a binary choice that serves neither side well.

The traditional route, photographers, videographers, models, and studios, is slow and expensive, and that cost structure is becoming harder to justify as AI tools mature.

The alternative many teams have adopted instead is a patchwork of five, six, sometimes ten separate AI subscriptions, each handling one piece of the workflow, none of them talking to each other.

Neither path solves the actual bottleneck, which is not generation capability. It is the cognitive and operational overhead of stitching together a coherent campaign from disconnected tools and disconnected outputs.

The founder is not selling image generation. He is selling the removal of a coordination tax that every team in this space currently pays, whether they have noticed it or not.

---

The Strategic Decision Layer

Avocado AI was originally built and marketed with a specific message aimed at e-commerce brands.

It was deliberately positioned as an alternative to hiring an agency, with early focus on the skincare sector and a strong early market in South Korea.

The go-to-market copy said, in effect, skip the agency, do it with Avocado. But the market answered differently.

The platform’s largest customers turned out to be the agencies themselves, the exact party the original positioning had implicitly tried to bypass.

Most founders would treat that as a messaging failure to be corrected. This founder treated it as signal and rebuilt the product’s central thesis around it.

Rather than fighting to convince agencies that AI replaces their function, Avocado AI repositioned around collaboration, a workflow where human judgment and AI generation operate together rather than one substituting for the other.

That decision shows up architecturally, not just rhetorically. The platform’s internal AI agent, Leni, is built with an explicit constraint the founder imposed himself, it must ask permission before publishing anything autonomously.

The founder is, by his own description, an extremely technical builder who can move fast.

The harder discipline was building a system capable of more autonomy than he was willing to grant it on day one, and choosing restraint anyway until trust in the output was earned through repeated testing on his own business first.

The decision to delay broader release of the most ambitious feature, the conversational super-agent capable of researching, drafting, and autonomously building campaigns, is also deliberate rather than resource constrained.

The founder has chosen to build it publicly alongside a small group of real users specifically so he could observe how people actually interacted with it before widening access.

---

Ecosystem Context

What this founder’s account of representation and access reveals is not a claim specific to one country or one sector.

It is a structural observation about how founders from underrepresented backgrounds calculate risk differently before they ever approach an investor or a market.

The founder describes the absence of visible Black founders who had achieved meaningful scale as a direct contributor to years of hesitation before starting Avocado AI.

This is not a complaint about a specific instance of discrimination. It is a description of how representation gaps function as a quiet tax on ambition, operating well before any formal barrier, pitch meeting, or funding decision enters the picture.

Founders who do not see people like themselves succeeding at scale carry a heavier internal burden of proof before they will take the leap at all.

That observation connects directly to a second one later in the conversation, about where founders choose to build.

The founder is candid that had he remained in Brazil, his most likely career trajectory by his current age would have been a comfortable, senior design leadership role inside an established company, not a startup.

He attributes the shift specifically to relocating to Estonia, describing an ecosystem where founder failure, restart, and eventual success are visibly normalised in a way he did not experience at home.

People around him try, fail, and try again openly, and that visibility changed what felt possible for him personally.

This is a meaningful ecosystem signal for any institution thinking about founders from Brazil and other emerging markets specifically.

The constraint is frequently not capability or ambition. Rather, it is the absence of a visible, normalised culture of entrepreneurial risk-taking close enough to a founder that failure stops looking catastrophic and starts looking like a normal part of the process.

While currently operating in Estonia and the broader EU, the founder identifies no current regulatory obstacle to what Avocado AI builds today, because the platform relies on existing third-party foundation models rather than training proprietary ones.

Though, he is explicit that building an in-house large language model or a proprietary image model might introduce a materially different and more complex regulatory relationship with GDPR.

---

Observed Patterns

The founder describes multiple prior products built but never shipped, held back specifically by fear that the work would not be good enough or would not sustain his interest over time.

One of those unreleased products shared its name and core concept with an application that someone else later built, released, and scaled to several hundred thousand dollars in monthly revenue.

The founder’s own framing of this which is, watching an idea he had already built sit unreleased while someone else captured the outcome, is a candid and unusually specific account of the cost of excessive caution at the release stage.

That history reframes the current discipline around Avocado AI’s super-agent rollout. The founder is not simply cautious by temperament.

He has direct, painful evidence of what overcaution previously cost him, and the current approach, controlled testing with real users before wider release, reads as a corrected version of a pattern he has already paid for once.

The most direct, founder-named mistake is a product development one.

Significant early building happened based on instinct rather than direct user input. This has produced what the founder himself calls a Frankenstein product, with usage data later confirming that large parts of it were barely used at all.

The corrective action, prioritising customer conversations and usage data over instinct-led building, is now explicitly part of how the founder is operating, including a self-imposed rule to pause further feature development until a defined number of customer conversations are completed.

---

Open Variables

Although the early monthly recurring revenue (MRR) from Avocado AI is quite decent the founder describes distribution as the single hardest current challenge, ahead of any technical constraint.

Whether the platform’s collaboration-first positioning and team-based pricing model will robustly compete against rivals on visibility alone, is yet to be determined still at an early stage.

The technical scaling threshold is a second open variable with a defined trigger point. The current infrastructure handles usage spikes without strain even though the founder understands the specific volume threshold at which load balancing and distributed infrastructure investment will become necessary.

This is linked directly to the most likely future trigger point for seeking external investment and does suggests that the company’s capital strategy is currently reactive to a technical threshold rather than a fixed timeline.

The relationship between the workspace product and the super-agent is not yet fully resolved in the public narrative given that the latter is still being tested.

---

Why This Matters

For founders in emerging markets building inside saturated global categories, this case offers a specific lesson about competitive moats.

Feature parity in generative AI tools is now achievable within days of any release, as this founder describes from direct experience watching competitors copy his work almost immediately.

In categories where the underlying technology is rapidly commoditised, the durable advantage increasingly sits in workflow design, pricing structure, and trust architecture rather than in the generation capability itself.

For ecosystem builders and policy makers in emerging markets and other comparable markets, it is worth noting the differences and outcomes of building within these arenas.

This is because the nature of visible, repeated public examples of founders failing and trying again appear to function as a structural enabler of risk-taking in a way that success stories alone do not replicate can be different within two different markets or geographies.

For accelerators and DFIs operating across emerging markets broadly, the distribution challenge, strong product and weak commercial visibility against better-funded competitors are just some of the most common and most addressable gaps in early-stage support.

Capital alone does not solve distribution. Founders with genuinely differentiated products but limited go-to-market reach are a category that targeted marketing or partnership support could disproportionately benefit.

---

Final Strategic Takeaway

One of the most important sentences in this entire conversation is not about the product but the environment within which a product is conceived and built

Avocado AI exists because the environment changed. Nevertheless, the product still has to win in one of the most contested categories in applied AI today, against competitors.

Avocado AI is also a story about how long capable people can sit on capability they already have, waiting for permission that was never going to arrive externally, until something in their environment makes starting feel survivable rather than reckless.

---

This article is drawn from an in-depth founder interview conducted by Afriq IQ with Wanderson Jackson, founder and CEO of Avocado AI. Selected insights and observations are published here.