Datadog CEO: AI won't stop us from writing code

Olivier Pomel, co-founder and CEO of Datadog, shared insights into the company's journey and his philosophy on leadership and technology at a recent Startup School event. He revealed that Datadog, which monitors the cloud infrastructure for numerous companies, has grown to encompass 20-25 products, driven by customer usage and strategic bets on future trends.

The AI Shift: From Writing Code to Automating It

Pomel addressed the growing influence of AI in software development, stating that his co-founder once told the engineering team they would stop writing code within two quarters. However, Pomel clarified that this doesn't mean an end to coding. Instead, he envisions an inversion: "You used to mostly write, sometimes automate, and now you're going to mostly automate, sometimes write." This shift, he believes, will empower smaller teams to achieve more, even if it means making more mistakes along the way.

The full discussion can be found on YC's YouTube channel.

The company has seen tangible benefits from this AI-driven transformation, with developers rebuilding entire systems in days that previously took teams months. This has convinced Datadog to pivot its organization, acknowledging that while the transition is painful, it's necessary for future growth.

From Rejection to Resilience: The Datadog Story

The conversation touched upon Datadog's early days, including their initial failure to get into Y Combinator. Pomel recounted receiving a rejection email from Paul Graham himself, a moment that fueled their motivation to prove their doubters wrong. This rejection, he noted, was a powerful impetus for the company's subsequent success.

The journey wasn't without its challenges. Pomel described the difficulty of raising their first angel round, facing skepticism from investors who didn't understand their market. He also highlighted the geographic challenges of pitching to VCs in New York and the Bay Area, who were initially hesitant to invest in infrastructure companies outside of their immediate locales.

Culture, Leadership, and Staying Grounded

Datadog's culture, according to Pomel, flows from the top, with leaders embodying the values and making hiring, promotion, and firing decisions based on cultural fit. He finds the notion of writing down culture on a wall unnecessary if the core principles are lived daily.

Pomel also emphasized the importance of management staying connected to the ground reality of the company. He reads employee survey comments and samples customer feedback, sales transcripts, and support requests to maintain a clear understanding of the company's fabric. This direct engagement allows him to identify issues that might be smoothed over in management summaries, ensuring that the company remains responsive to customer needs and market dynamics.

Navigating the Future: AI, Public Markets, and Expansion

Looking ahead, Pomel acknowledged the rapid pace of change, particularly with AI, and the need for companies to be comfortable with making mistakes. He identified investing in product capabilities for AI agents and building intelligence directly into their observability platform as key bets for Datadog.

He also discussed the impact of being a public company, noting that while it requires more choreographed communication, it doesn't fundamentally change the core leadership approach. The key, he suggested, is to consistently tell the right story and produce numbers that align with that narrative.

When asked for advice for aspiring entrepreneurs, Pomel stressed the importance of moving faster, especially in hiring and firing decisions. He noted his own past paralysis in trying to make the perfect hire, a mistake he now believes is counterproductive. "It's much better to do that than to wait... you'll have more cycles, you have more chances," he advised.