What is Symflower?

Symflower helps you build better software by pairing static, dynamic and symbolic analyses with LLMs. The robustness of deterministic analyses combined with the creativity of LLMs allows for higher quality and rapid software development.

Works where you are:

• In your console and CI via CLI
• As component in your application and SOA via LSP
• In your VS Code-based editors: VS Code and Cursor
• In your IntelliJ platform IDEs: IntelliJ IDEA, Android Studio, GoLand
• As a drop-in proxy for LLM inference endpoints

Make LLMs better

• Find the best and cheapest model for your programming language, framework and use case
• Create evaluations for your environments, workflows and requirements
• Run benchmarks continuously to make sure your evaluation still works and supports the latest models
• Automatically augment your LLM requests (RAG) for better results
• High-quality training data curation for more powerful models
• Automatically fix and improve generated code using pre- and post-processing
• Flag and fix low-quality generated source code
• Help LLMs generate usable tests via smart test templates
• Automatically run just the right tools for your environment
• Rapid testing of LLM-generated changes via test impact analysis
• Fine-tuning for your DSL (domain specific language)
• Track LLM-generated changes in code bases for liability, security and compliance
• LLM inference endpoints in your own infrastructure to eliminate external dependencies

Make your editor better

• Generate smart test templates with imports, annotations, object initializations, function calls, asserts, and more
• Automate the generation of complete test suites via symbolic execution or via LLMs
• Post-processing with linting and code repair to improve code quality
• Efficiently create, duplicate, and review tests
• Automatically identify uncaught exceptions in your source code in real time with tests to reproduce them

Make your development process better

• Programming language-agnostic tooling for a streamlined development workflow
• Rapid testing both locally and in your CI with test impact analysis
• Static analysis and automated fixes to improve low-quality and erroneous code
• Automatic code reviews and fixes to ensure high code quality
• Source code transpilation across programming languages
• Migrating to other programming languages, frameworks and libraries
• Analyze runtime performance in sequential and concurrent applications

Can you think of additional tooling that would improve your daily (LLM-assisted) software development life? Let us know how we can help you!

Feedback

To provide feedback about features you want to see and problems you have discovered, please use our community issue tracker. For general discussions and questions, use the community discussion board.