Nana Muazin
In a landmark move for the artificial intelligence industry, Cognition, the high-profile startup behind the AI coding agent Devin, has announced the acquisition of The Interaction Company of California, the creators of the popular conversational AI assistant, Poke. The deal, which industry insiders confirm is valued in the “low nine figures,” signals a pivotal shift in how AI developers are prioritizing human-computer interaction (HCI) alongside raw computational power.
By integrating Poke’s signature conversational personality with the technical prowess of the Devin coding agent, Cognition aims to move beyond the traditional "tool-like" experience of software, aiming instead for a future where AI feels less like a sterile program and more like a proactive, witty, and reliable human colleague.
The Main Facts: A Meeting of Minds and Models
The acquisition represents more than just a merger of two tech companies; it is a synthesis of two distinct philosophies in AI development. Poke, which launched in March 2026, quickly gained a reputation for being the AI assistant you text like a friend. Unlike traditional assistants that provide cold, transactional responses, Poke was designed to leverage slang, humor, and a deep understanding of natural human cadence.
Cognition, meanwhile, has been laser-focused on the technical heavy lifting of software engineering. Their flagship product, Devin, has been widely touted as the world’s first autonomous AI software engineer, capable of handling complex coding tasks, bug fixes, and system deployments.
The deal will see the expertise of The Interaction Company team folded into Cognition’s operations. While Poke will continue to function on existing platforms—including its pioneering role as the first AI agent approved for Apple’s Messages for Business—it will gradually transition to utilizing Cognition’s advanced infrastructure and the new SWE-1.7 software engineering model.
A Chronology of Growth and Innovation
The trajectory of both companies over the past year has been nothing short of meteoric.
- March 2026: The Interaction Company launches Poke, introducing a novel way to interact with AI via standard messaging channels like SMS, Telegram, and iMessage. The platform experiences immediate viral growth as users lean into its conversational, personalized nature.
- April 2026: TechCrunch profiles Poke, highlighting its ease of use and unique position as an "AI friend."
- June 2026: In a significant industry milestone, Poke receives official approval from Apple to operate on its "Messages for Business" platform, cementing its legitimacy as a enterprise-grade communication tool.
- Late Q2 2026: Cognition, already deeply familiar with the team’s work—partly because co-founders Scott Wu and Walden Yan were early angel investors in the venture—initiates formal acquisition discussions.
- Current Day: The acquisition is finalized, setting the stage for a hybrid future where coding agents become inherently more social and personality-driven.
Supporting Data: The Scale of the Engagement
The strength of the acquisition lies in the sheer volume of interaction data Poke has cultivated. In just three months of operation, users exchanged more than 100 million messages with the Poke AI agent. This figure is staggering, considering the platform serves hundreds of thousands of users who rely on it for diverse needs, ranging from travel planning and financial tracking to scheduling and educational queries.
However, the rapid growth also exposed the challenges of the current AI economic model. As co-founder Marvin von Hagen candidly admitted, maintaining such a high level of conversational quality at scale is computationally expensive. Despite the high engagement, the operational overhead made achieving profitability a difficult hurdle. By moving under the umbrella of Cognition, Poke gains the infrastructure and capital necessary to scale its technology sustainably while maintaining the "personality" that attracted its massive user base.
Official Perspectives: Why Personality Matters
The rationale behind the deal was articulated clearly by Cognition’s leadership. Scott Wu, co-founder of Cognition, emphasized that the goal is to bridge the gap between "software" and "colleague."
"The Interaction team has built an agent that people love: it’s proactive, it knows you, and it’s fun to talk to," Wu wrote in an official company blog post. "That’s exactly how working with Devin should feel, and now we get to build it together."
Marvin von Hagen echoed this sentiment, arguing that the future of work is not just about the efficiency of the output, but the nature of the interaction. "You probably prefer it if you have co-workers that have personality, rather than if you have co-workers that are just robots," von Hagen said in an interview. "When you have co-workers that are also software engineers, they can also make a joke. And you’ll find it funny and enjoyable."
This philosophy underscores a growing consensus in Silicon Valley: as models become more capable, the differentiator will no longer be "intelligence" in a vacuum, but the "user experience" of the collaboration.
Implications for the Future of AI Agents
The acquisition of Poke by Cognition is likely to have long-lasting ripple effects across the AI sector.
1. The Rise of Persistent, Proactive Agents
One of the primary limitations of current coding agents like Devin is their tendency to be ephemeral; they operate on a per-task basis. By incorporating Poke’s interaction model, Cognition is looking to create a "persistent co-worker." This means an agent that remembers your preferences, your previous coding styles, and the context of projects that span weeks or months.
2. Orchestration as the Next Frontier
Von Hagen suggested that Poke’s future role might involve "orchestrating" Devin’s sessions. Currently, Devin is excellent at executing a specific pull request, but it lacks the oversight to manage complex, multi-layered projects across different sessions. Poke could serve as the "manager" or "orchestrator," keeping the coding agent focused, on track, and aware of the broader goals of the software development lifecycle.
3. The "Personality" Premium
This acquisition validates the belief that personality is a value-add, not just a gimmick. As AI becomes ubiquitous, developers who can successfully imbue their models with a sense of "self" or "identity" will likely see higher retention rates and better user satisfaction. The ability to handle complex technical tasks while maintaining a casual, friendly, and even humorous tone is the next benchmark for AI companies.
4. A Template for Future Consolidations
We should expect to see more "UI/UX-first" AI companies being acquired by "Model-first" AI companies. Startups that have mastered the art of natural conversation—but struggle with the high costs of infrastructure—are prime targets for larger labs that have the compute power but lack the consumer-facing polish.
Conclusion: A New Standard for Collaboration
As we look toward 2027 and beyond, the line between software and colleague will continue to blur. The Cognition-Poke merger is a bold bet that the future of engineering is not just about writing code faster, but about building systems that we actually want to work with.
For now, users of Poke will see little change as the team spends the remainder of the year integrating their workflows. However, by next year, we can expect a new iteration of Devin—one that is perhaps a bit more "human" in its approach, more reliable in its memory, and infinitely more pleasant to talk to when the code inevitably breaks.
In a world where AI is becoming the primary interface for our digital lives, Cognition is betting that the most successful systems will be the ones that know how to keep it professional—while still knowing how to tell a joke.
Ammar Sabilarrohman
Pevita Pearce
