Building an AI-First Company: Aircall's Journey with Generative AI

Ready to build better conversations?

Simple to set up. Easy to use. Powerful integrations.

Get free access

As Large Language Models (LLMs) rapidly evolve with new capabilities, companies across the tech industry are exploring how to leverage this transformative technology—both to boost internal productivity, and to enhance their products. At Aircall, we’re facing the same exciting challenge and have carefully mapped out our strategy to use Generative AI (GenAI) as a key driver of competitive advantage.

Democratizing AI Across Aircall

We believe that empowering every department to incorporate GenAI into their workflows is the key to unlocking its full potential. Instead of a top-down approach, we’ve embraced a ‘democratized’ model, enabling employees across all functions to experiment with GenAI tools, share insights, and build new capabilities.

We kicked off our journey by piloting a simple Confluence-based knowledge agent, designed to help our teams quickly access internal knowledge. As we continue to refine this solution, we’re building training materials to onboard employees from all corners of the company, giving them the tools they need to create and interact with their own knowledge agents. We’ve also partnered with external AI providers to run training sessions, igniting curiosity and excitement about how GenAI can revolutionize our workflows.

Looking ahead, we’re planning internal hackathons focused on GenAI, where teams can put their creative ideas into action. These events will give everyone the chance to explore GenAI’s potential, share learnings, and build innovative solutions that can benefit the entire company.

Guiding Principles for GenAI at Aircall

As we develop our approach to building with GenAI, a few key principles are shaping our journey:

1. Learning by Doing Sparks Innovation

The more our teams engage with GenAI, the more ideas we generate for its application. GenAI is not only limited by the scope of our imaginations but also by the capabilities of the models themselves. We believe that innovation happens through hands-on experimentation, which is why we’re encouraging a culture of building and iterating with GenAI.

2. Data Defines AI Success

While many companies use GenAI to navigate internal knowledge bases, the results vary significantly depending on how their data is structured. Larger organizations often have fully integrated tech stacks, but as a rapidly growing company, Aircall relies on a variety of third-party tools such as Confluence, Git, and Google Docs. This means that our knowledge is spread across multiple platforms, and unlocking the power of GenAI depends on how well we can integrate these disparate data sources.

3. Product Teams Drive Customer-Focused AI Innovation

Our product teams know our customers and their pain points best, making them the ideal drivers of GenAI innovation. While we have a dedicated AI team working on core capabilities, we’re empowering every product  and engineering leader to explore how GenAI can improve their features and deliver a better experience to our users. Our goal is to foster a mix of internal productivity improvements and customer-centric AI features that can eventually be productized.

For instance, if we develop an AI-driven tool to enhance the efficiency of our customer support team, we could later offer that same solution to customers using Aircall to streamline their own support operations. This dual approach—focused on internal gains and customer value—will help us continuously iterate and innovate.

4. Leadership Support is Critical to Becoming an AI-First Company

Our leadership team is committed to making Aircall an AI-first company. They actively participate in AI-related initiatives, from contributing to internal Slack channels to reviewing demos and offering strategic input on how we prioritize AI development.

As we navigate key decisions—such as whether to prioritize building general-purpose AI agents or specialized, highly accurate ones—our leaders provide guidance to ensure we strike the right balance between innovation and reliability.

“We believe in a product that's people-first. We see AI as a way to let you do less through automation, and to focus on doing the most critical things even better than before.”

• Scott Chancellor, Chief Executive Officer, Aircall

5. Cross-Functional Collaboration is Essential

Building with GenAI isn’t just a technical challenge; it requires cross-functional collaboration between teams like legal, security, and operations to ensure our AI solutions are safe, ethical, and reliable. We are proactively developing guardrails to prevent misuse, such as prompt injection attacks, and ensuring that any AI-driven features we launch are built with input from across the organization.

6. Establishing an Internal AI Playbook

As LLM capabilities evolve rapidly, so do the best practices for using them. To stay ahead, we’re creating a GenAI playbook—a living document that captures our learnings, experiments, and the latest AI strategies. This playbook will enable us to quickly share knowledge across teams and continuously refine our AI development approach as new tools and models emerge.

Looking to the Future

Aircall’s GenAI journey is just beginning, but we’re already seeing the impact of a democratized, hands-on approach. By empowering everyone to experiment with AI, encouraging cross-team collaboration, and maintaining strong leadership support, we’re laying the foundation for a future where GenAI drives both internal efficiencies and exciting new product features for our customers.

Stay tuned for more updates as we continue to explore the possibilities of GenAI and build innovative solutions that push the boundaries of what’s possible at Aircall.

Want to learn more about our innovations and what goes on in the engine room of Aircall? Take a look at our Tech Team Stories.

Published on December 4, 2024.