Made By
bloopReleased On
2021-10-24
Bloop is an AI-powered software tool designed to modernize legacy code, particularly COBOL, by converting it into modern Java code while maintaining identical functionality. This tool assists development teams in updating their codebases, writing new code, and understanding complex software systems through natural language interactions.
Key features:
- Automatic Code Conversion: Transforms legacy COBOL code into immediately modifiable and maintainable Java code.
- Behavior Preservation: Utilizes an AI test suite to ensure converted Java programs exhibit identical behavior to the original COBOL code.
- Readable Output: Produces Java code that is as readable as if written by a human, avoiding hybrid language issues.
- Open-Source Utilization: Employs open-source libraries and cloud primitives in modernized code to maximize cost savings.
- Immediate Modification: Allows for code modification and extension post-conversion without the need for refactoring.
- Continuous Development: Enables ongoing project progress without requiring code freezes during the conversion process.
- AI Pair Programming: Incorporates mAInframer-1, an AI model trained specifically for COBOL developers to enhance productivity.
- Offline Functionality: Runs the AI model entirely on the user's device, eliminating the need for an internet connection.
- Natural Language Queries: Supports asking questions about the codebase in natural language for improved understanding.
- Multi-Language Code Navigation: Facilitates navigation to function definitions and usage examples across multiple programming languages.
- Multi-Repository Search: Comprehends relationships between different codebases and service interactions.
- Regex Matching: Includes a purpose-built regex engine for matching code, identifiers, and paths.
- Code Diffing: Enables rebuilding or rewriting code while preserving its scope and specifications.
- Documentation Integration: Allows importing of documentation libraries to assist in code writing.
How it works:
1. Users upload their legacy COBOL code through the web app or IDE extension.
2. Bloop converts the uploaded code into modern Java code.
3. Developers can use the AI pair programmer to write new COBOL code or modify existing code.
4. Users can ask questions in natural language to better understand the codebase and navigate through it.
5. The tool provides regex matching capabilities for specific code searches.
Integrations:
Bloop supports integrations with various development tools and platforms.
Use of AI:
Bloop uses generative AI to facilitate the code conversion process and to power its AI pair programmer feature. The AI also enables natural language interactions for code understanding and navigation.
AI foundation model:
Bloop utilizes a large language model called mAInframer-1, which is specifically trained for COBOL developers. This model is built on open-source code and designed to run offline to ensure privacy and security.
Target users:
- Development teams working with legacy code, especially COBOL
- Organizations seeking to modernize their codebases without extensive refactoring
- Developers looking to improve productivity with AI assistance
- Teams needing to efficiently understand and navigate large, complex codebases
How to access:
Bloop is available as a web application and potentially as an IDE extension.
Open Source Status:
While Bloop uses open-source code for training its AI model and produces modernized code that utilizes open-source libraries, the tool itself is not explicitly stated to be open source.
Pricing model: Unknown |
No hype. No doom. Just actionable resources and strategies to accelerate your success in the age of AI.
AI is moving at lightning speed, but we won’t let you get left behind. Sign up for our newsletter and get notified of the latest AI news, research, tools, and our expert-written prompts & playbooks.
AI is moving at lightning speed, but we won’t let you get left behind. Sign up for our newsletter and get notified of the latest AI news, research, tools, and our expert-written prompts & playbooks.