Introducing poolside's Assistant powered by malibu and point models
poolside addresses the challenges of modern software engineering for large enterprises. poolside's Assistant, powered by its generative models malibu and point, excels at code generation, testing, documentation, and real-time code completion with advanced context awareness. AWS will soon be the first cloud provider to offer fully managed models from poolside.
poolside's Assistant
poolside's generative AI Assistant puts the power of poolside's malibu and point models right inside of your developers' Integrated Development Environment (IDE). poolside models are fine-tuned on your team's interactions, code base, practices, libraries, and knowledge bases.
malibu
poolside's malibu model excels at tackling complex software engineering challenges, such as code generation, test writing, refactoring, documentation, and more.
point
poolside's point model excels at low-latency code completion and leverages advanced context awareness to accurately predict developers' needs.
Benefits
Meet poolside
poolside is pioneering the next generation of AI-powered software development tools, purpose-built for modern engineering teams. poolside’s foundation models (FMs), malibu and point, are trained using Reinforcement Learning from Code Execution Feedback (RLCEF), a novel approach to building AI models that improves performance in coding tasks. poolside is deployed in your environment, which allows you to securely and privately connect it to your data and tailor it to your needs—ingesting your code bases, documentation, and knowledge bases to create a model that is uniquely suited to your dev teams and business.
Use cases
Model versions
malibu (coming soon)
Excels at tackling complex software engineering challenges, such as code generation, test writing, refactoring, documentation, and more.
Max tokens: 100,000
Languages: English
Fine-tuning supported: Yes
Supported use cases: Code generation, software engineering challenges, test generation, documentation generation, custom and fine-tuned development.
point (coming soon)
Engineered for rapid code completion, leveraging advanced context awareness to accurately predict developers' needs.
Max tokens: Over 1 million
Languages: English
Fine-tuning supported: Yes
Supported use cases: Rapid code completion, context-aware predictions, in-line API suggestions, and syntax error prevention.