Three years ago, my co-founders and I started Faros AI, with the vision of making every company a world class software company. Our background was in building machine learning products and engineering teams, and we were motivated by our frustration with the complete black-box that is engineering operations today.

For context, we were developing the Einstein Machine Learning Platform at Salesforce. We found that while we were helping Salesforce customers harness AI to improve business outcomes, our visibility and insight into our own engineering processes was sorely lacking. And it wasn’t just us. Most sizable software engineering organizations today are largely flying blind.

Faros means lighthouse in Greek, and we called our company Faros AI, inspired by an ongoing nautical theme for Dev tooling (Docker/Kubernetes etc.), as well as the vision of helping engineering teams smoothly navigate troubled waters by shining a light on their operational bottlenecks and hotspots.

Now, there’s two possible ways to build an AI company. You either build the AI, and then look for data. Or you start with the data, and then build the necessary AI. We chose the latter:

Software engineering organizations typically use many dozens of systems to manage their engineering processes — from issue management, to continuous integration and delivery, to cloud infrastructure operations, budgeting, procurement, HR operations, and more. For the most part, none of these systems talk to each other or to any central system, yet many of the questions that engineering organizations need to answer involve querying data across these different sources.

Our focus since inception has been to build out a solid data foundation for all this data, with integrations to every engineering system out there - whether vendor or home-grown; standardization of a single, connected data schema to represent the entire SDLC; and layering of capabilities for cataloging, analytics, and automation.

But the volume of data flowing through engineering organizations is simply massive, and the sheer number of metrics and insights to be derived from it can be overwhelming. With the advent of large language models (LLMs), there’s never been a better time to harness AI to solve this problem. Today we are very excited to reveal Lighthouse AI - the foundational artificial intelligence engine built into the Faros platform, designed to help engineering organizations make sense of the vast amounts of data that they produce every single day.

The initial push in our AI strategy is on the following fronts:

1. Natural language based data exploration: One of the key challenges with data analysis for engineering operations is not just the volume of the data, but also the complexity. The software development life cycle is complex, the schema to represent it - even more so. Teams would typically need to hire trained data analysts, deeply familiar with both the data and the teams’ processes (with all their quirks) to translate business questions into performant and accurate SQL queries and dashboards. With the advent of LLMs, this should be a thing of the past. With Lighthouse AI, an engineering leader will be able to ask Faros for the most interdependent teams in their organization, the biggest bottlenecks in their application lead times, and the distribution of code review load across teammates, correlated with seniority. All this, in plain English, without the need for a deep understanding of the ins and outs of the underlying schema.
   
   Our goal with Lighthouse AI is to make querying operational data as simple as possible, so that every user of Faros can be a power user.
   
2. Guided navigation: Lighthouse AI will also change the way users navigate through our data products. Instead of static dashboards, AI algorithms will sift through the data, identify trends, highlight anomalies, and suggest areas of focus. Machine learning models will alert on issues before they disrupt operations, and correlate signals from disparate systems to help in causal analysis.
   
   In short, Lighthouse AI will tell Engineering teams what they need to care about, when they need to care about it.

The AI revolution has only just begun. We anticipate that every aspect of software engineering is going to be transformed by AI in the next five years, and at Faros, we are making sure that operational intelligence keeps pace, allowing engineering organizations to make that transition with confidence.

Interested in learning more? Request a demo and we will be happy to set up time to walk you through the platform and Lighthouse AI.