Delta-v partner Dan Williams recently sat down with Matt Arbesfeld to discuss how his company, LogRocket is leveraging AI/ML. Matt is the Co-Founder and CEO of LogRocket, a session replay, product analytics, and error-tracking tool that empowers software teams to create the ideal product experience. LogRocket’s AI tool, Galileo, leverages the power of artificial intelligence to help companies identify and resolve the most impactful areas for improvement in their software. Matt discussed the decision to build an AI/ML application, the process for building it, and lessons learned along the way.
Here’s an executive summary of the discussion, please reach out to your favorite Delta-v contact if you are interested in reading the full piece.
LOGROCKET’S AI OPPORTUNITY
LogRocket helps companies identify and correct usability and performance issues with their software products. A well-designed AI augmentation could sort through many alerts to focus on high-priority issues.
THE BUSINESS CASE FOR AI
The idea for an AI application came from challenges faced by clients who needed to sort through a high volume of alert data. LogRocket was able to validate that AI filtering would drive customer value and build a model.
BUILDING AN AI APPLICATION
Any purpose-built AI application will require customization, but you don’t need to (and shouldn’t) build a model completely from scratch. Nor do you necessarily need to hire specialized AI engineers, you can empower smart engineers on your team who have an interest in machine learning and data science experience.
OPTIMIZATION AND ACCURACY
Tuning your model should move through three stages: first, customer pain point identification, second, beta model, and third, a fully-automated model. Set a threshold at which the precision of the results from the model is acceptable to decide when it’s ready for customers.
Are you using AI/ML to enhance your product offering? We’d love to hear about it. Email Dan at email@example.com