Most pre-seed investors say they back founders. Far fewer can explain what that actually means when there is no product, no revenue and no proof the company will work.

That question sat at the heart of an EUVC conversation hosted by Andreas Munk Holm with Pooné Mokari, CEO & Co-Founder at Ewake, and Pietro Bezza, Co-Founder and Managing Partner at Connect Ventures.

Pooné described modern engineering environments as overwhelmed by operational noise, where developers must manually correlate “logs, metrics, code changes” and fragmented infrastructure data just to understand what is happening during incidents. Paris-based Ewake is building AI agents for software production reliability.

Existing observability platforms already provide visibility into systems, she said, but not understanding. Engineers still need to sift through huge amounts of context manually while incidents continue costing companies money.

“Our mindset was building a kind of operator,” Pooné said. “We call it a teammate.”

For Pietro, the product itself was only part of the investment case. Connect Ventures invests primarily at pre-seed, often before products exist. Pietro said that at that stage, “everything that we assess are the founders.”

He described Pooné and her co-founder as “great archetypes and examples of the product founders” Connect looks for.

Pietro said Connect’s core belief is that “the best technology companies are product companies.”

According to Pietro, product-first founders think differently from marketing-led or distribution-led founders because they focus obsessively on creating conditions for great products to emerge.

“Great products don’t grow on trees,” Pietro said. “You create and foster and design the condition for great products to be built.”

One of the strongest signals for Connect, he said, is whether founders have deep and opinionated insight into an unsolved problem.

Ewake stood out because Pooné and her co-founder had lived the reliability problem themselves during their time at Criteo, where production uptime was mission-critical.

Pietro said the team combined “insider’s view into the observability world” with “an outsider point of view on how to solve it.”

For Pooné, the motivation came directly from operational pain.

She described engineers as flooded with reactive observability systems that force teams to predefine what they monitor and alert on. When unexpected incidents happen, she said, companies often discover they never instrumented the right thing.

When asked about raising a €2 million pre-seed round before having a finished product, Pooné immediately returned to the problem itself rather than fundraising tactics.

“If you are solving a problem, you know why this problem exists and where it exists,” she said.

Pooné said the Ewake team already knew where the product should live operationally because they had experienced the workflow directly. Engineers were already coordinating incidents through Slack, she explained, so building another standalone dashboard made little sense.

“If I wanna have a teammate, it shouldn’t be just opening another tab,” Pooné said.

Pietro explained why Connect prioritises founder insight over operational track record at pre-seed.

Execution capabilities can eventually be hired, he said, but authentic problem understanding cannot.

That philosophy also shapes how Connect thinks about repeat founders. Experience is valuable, Pietro said, but it can sometimes create biases around distribution, go-to-market playbooks and scaling motions that pull founders away from product obsession.

For him, product-first thinking is also distinct from product-led growth. “Product-led growth is how you grow your business,” he said. Product-first thinking instead shapes how founders allocate capital, attention and resources across the company over time.

The strongest companies continue reinvesting into product development long after initial success. Companies that failed to grow beyond “50 million or 100 million in revenues,” Pietro said, often struggled because “they failed to build product two.”

For Connect, product obsession is not just an early-stage signal. It becomes the long-term operating system of enduring companies.

“The great founders, the best product companies,” Pietro said, “always find a better way, a new way to solve the problem better.”

The conversation also highlighted why operational AI products require a different approach from many consumer AI applications.

Pooné said Ewake is heavily focused on design partnerships because AI reliability systems only work if engineers trust them during critical incidents.

“A real success for a company like Ewake is they trust Ewake during critical moments,” she said.

That requires staying extremely close to users. Pooné described the company as obsessed with feedback loops, user workflows and developer behaviour, with Ewake even using its own product internally to evaluate how engineers interact with the system during incidents.

“Being very focused on the users, I think, is the most important step in building a great product,” Pooné said.