FAQs
What does petals.ai do?
Problem
Legacy systems, systems that are built on old technologies, are hard to maintain and evolve. Companies often lack the documentation and knowledge to understand how these systems work. This becomes a bottleneck when you want to add new features or modernize the system.
Our Solution
Petals.ai has an Agent AI powered tool that extracts critical business knowledge from legacy applications. Customers can feed this generated documentation to AI based IDEs (like GitHub Copilot, Amazon CodeWhisperer, etc) to generate modernized applications or implement new packaged applications.
When should you engage with petals.ai?
Following are the 6 R's of application modernization:
- Rehost. Often referred to as lift and shift, This strategy does not modifications to application's code.
- Replatform. Sometimes called lift, tinker, and shift, this strategy moves the application to a new runtime platform with minimal code changes.
- Refactor. This strategy involves changes to existing code without major changes to an application's external behavior.
- Rebuild. Starting over and Rebuilding entire application
- Retire. Decommissioning or shutting down applications.
- Retain. Do Nothing
Petals.ai focuses on the refactor, rebuild and retain strategies. We will not be of much help in the rehost, replatform or retire. If you decide to retain, Petals can significantly reduce the pain of maintaining applications.
How does petals.ai work?
- Rosetta (Petals.ai's Core Bot) connects to your source code repository to understand the business logic.
- Generates Extensive functional test cases that cover various scenarios and edge cases.
- Customer business analysts go through the generated test cases and validate them.
- Once all the test cases are identified and validated, these test cases will be the input for developing the new system.
Benefits of using Petals.ai
- It is estimated that the petals.ai bot will extract 50-75% of the test cases.
- Reduced dependency on subject matter experts. Save 50%+ time of experts
- Reduced onboarding time of new employees, from weeks to days
- Functional test cases are objective, measurable and can be validated by business.
- Impart your business domain knowledge to AI and reduce people dependency
- Test cases provide easier business process re-engineering
- Test cases are excellent input to generate Code using AI based IDEs like GitHub Copilot, Amazon CodeWhisperer, etc.
Petals.ai provides 2 key components. 1. A core engine bot that generates documentation 2. A Q&A Bot that acts as a guru on your legacy system answering questions. Following are the benefits of using Rosetta(Our Product)
Why can't I use chatGPT or other LLMs to generate test cases myself?
- The code base of legacy systems can be huge (millions of lines of code). It is not possible to feed the entire code base to LLMs due to token limits.
- You can not 'Teach' your company specific business knowledge to LLMs using chatGPT.
- Code has dependencies. This can not be captured by ChatGPT.
- You will need to do significant amount of prompt engineering to get the desired results.
About Petals.ai
Petals.ai is a product company based in Chicago, IL. We are a team of passionate engineers and entrepreneurs with deep expertise in AI, legacy modernization. Our mission is to help companies modernize their legacy systems using AI and reduce their dependency on people.
Though our solution is horizontal, (Works for all industries), we have deeper expertise in modernizing Supply Chain, Retail and Manufacturing industries.