SunRun taps into the power of conversation AI with Amazon Q Developer Slackbot

SunRun, founded in 2007, is a leading company in the residential solar energy sector. They offer homeowners various options to adopt solar power, such as leasing and purchasing solar panel, battery, and EV systems. SunRun became the first and only solar-plus-storage company in the U.S. to surpass one million customers.

As SunRun’s business has expanded rapidly, their engineering and cloud operations teams have faced increasing demands. To address these challenges and improve their workflows, SunRun implemented an innovative solution: they created an internal Slack workspace powered by Amazon Q Developer, a generative AI-powered assistant for software development.

This post examines how SunRun used Amazon Q Developer to integrate its chat capabilities using AWS Chatbot with its enterprise Slack workspace. We explore how this generative AI powered tool has significantly enhanced SunRun’s cloud operations and streamlined their development processes. Using Amazon Q Developer’s generative AI capabilities has allowed SunRun to manage their growing technological needs more efficiently while continuing to lead in the renewable energy market.

The challenge

SunRun sought to address two primary challenges:

1. Improving development velocity: SunRun needed to significantly boost the productivity of their developer and engineering teams. The goal was to speed up the entire process of creating and launching new products and services for their customers. This involved streamlining workflows, optimizing coding practices, and enhancing collaboration among team members to reduce development cycles and bring innovations to market faster.
2. Reducing dependence on AWS Support: SunRun aimed to reduce their reliance on Amazon Web Services (AWS) support services for questions on cloud infrastructure.

Addressing these challenges allowed SunRun to become more self-sufficient in managing their cloud environment, thereby improving their operational efficiency and reducing delays in problem-solving and decision-making processes related to their AWS infrastructure. “Decreasing the barrier to information was important to us. It’s much easier to type into Slack than to dig through online AWS sites for an answer” said Lee Granas, Sunrun Infrastructure Team Lead.

Solution overview

SunRun conducted a review of multiple generative AI powered assistants. As a result, they chose Amazon Q Developer, a generative AI-powered assistant trained on more than 17 years of AWS knowledge and experience to enhance their engineering team’s efficiency.

Amazon Q Developer can help a developer understand, build, transform, and operate AWS applications. The developer can ask Amazon Q about AWS architecture, AWS resources, best practices, documentation, support, and more. Amazon Q is constantly updating its capabilities, so the asked questions get the most contextually relevant and actionable answers. When used in an integrated development environment (IDE), Amazon Q Developer helps with all development tasks. Amazon Q Developer can chat about code, provide inline code completions, generate new code, test code, scan code for security vulnerabilities, and help quickly remediate them. Furthermore, Amazon Q can perform code upgrades and improvements, such as language updates, debugging, and optimizations.

Amazon Q Developer is powered by Amazon Bedrock, a fully managed service that makes foundation models (FMs) available through an API. Amazon Q uses different models for different workflows because it optimizes for latency, cost, accuracy, and reasoning capabilities depending on the task at hand. Amazon Q has been augmented with high quality AWS content to give more complete, actionable, and referenced answers to accelerate building on AWS.

The SunRun team developed a proof of concept to assess how well Amazon Q Developer’s recommendations perform in terms of accuracy regarding various aspects of AWS resource management and document reference. Integrating Amazon Q Developer into their enterprise Slack messaging workspace using the AWS Chatbot integration, SunRun empowered their teams to use the knowledge and capabilities of Amazon Q Developer directly within their Slack workspace without having to leave their primary communication platform. Upon successful completion of the POC, they found that the recommendations from Amazon Q offered:

1. Best practices guidance: provided detailed advice on optimal approaches for designing and implementing AWS solutions across various services and architectures.
2. Troubleshooting assistance: Amazon Q Developer offered step-by-step diagnostic procedures and potential solutions.
3. Next steps recommendations: based on the problem statement submitted, Amazon Q Developer suggested logical next actions, helping to streamline the development process.

Benefits observed by Sunrun:

Integrating Amazon Q Developer generative AI assistant with Slack allowed SunRun to empower its employees to self-serve on a wide range of cloud-centric queries and workflows. Developers could chat with the generative AI assistant in natural language, ask questions about best practices, and retrieve the latest AWS white papers/documentation. This integration proved to be a powerful tool for the company’s workforce, particularly its development team, offering several key benefits:

1. Self-service information access:

• Employees gained the ability to quickly retrieve information on a wide range of cloud-related topics.
• The generative AI assistant served as a centralized knowledge base, reducing the need for manual searches across multiple resources.

2. Natural language interface:

• Users could interact with the generative AI assistant using natural language.
• This ease of use encouraged more frequent engagement with cloud resources and best practices.

3. AWS best practices on-demand:

• Developers could instantly query the generative AI assistant for recommended approaches to cloud architecture, security, and performance optimization.
• This feature helped maintain consistency in cloud implementations across projects and teams.

4. Access to latest AWS documentation:

• The integration provided quick retrieval of up-to-date AWS white papers and documentation.
• This made sure that the development team always had access to the most current information on AWS services and features.

5. Improved efficiency:

• Reducing the time spent searching for information or performing routine tasks allowed developers to focus more on core development work. The generative AI Assistant served as a force multiplier, allowing the team to accomplish more with existing resources.

“One nice thing about integrating Amazon Q with Slack has been how quickly we can get answers. Instead of rifling through online documentation, I can just drop a question in the Slack channel and immediately get what I need.” – Lee Granas, Infrastructure Team Lead

“It’s great that we can ask questions as though we were talking to another person. It’s a lot faster and easier to get information.” – Andy VanSickle-Ward, Sr. Manager, Software Engineering

Conclusion

SunRun’s implementation of the Amazon Q Developer generative AI assistant in Slack has definitively improved SunRun’s development team’s interaction with AWS resources. This innovative solution has created a direct, on-demand channel for accessing AWS expertise, resulting in several significant benefits, such as increased development velocity, enhanced agility, and improved responsiveness.

Looking to the future, SunRun has aspirational plans to explore Amazon Bedrock generative AI platform capabilities:

1. Integration with internal systems: connecting the generative AI assistant with other internal platforms, creating a more comprehensive and interconnected information ecosystem.

2. Exploring new use cases: investigating how the conversational generative AI assistant can be applied to other areas of their business, such as:

• Customer service: potentially using the generative AI assistant to provide faster, more accurate responses to customer inquiries.
• Sales enablement: exploring ways that the generative AI assistant could support the sales team with real-time access to product information, pricing, and customer data.

Continuing to innovate with AI-driven solutions such as Amazon Q Developer allows SunRun to position itself at the forefront of technology adoption in the energy sector, making sure it remains adaptable and efficient in meeting the challenges of a dynamic market.