SajiCode - AI Engineering Teams

SajiCode is an AI engineering team that works directly in your terminal. It uses a team of 17 specialized agents, each with expert skills, to build software that can be used in real products. Instead of using just one AI, SajiCode works like a real engineering team with different people handling different jobs. This helps avoid common problems like losing track of big changes, writing unfinished code, skipping quality checks, and needing constant attention.

Benefits

SajiCode helps build software more efficiently by breaking down tasks among specialized AI agents. This multi-agent approach allows many tasks to be worked on at the same time. It also has built-in checks to prevent unfinished code, stop harmful actions, and avoid repeating tasks, which leads to better quality and safer code. Users can choose how much they want to be involved in critical decisions, like approving commands. SajiCode can remember past work and user preferences, making future tasks smoother. It can also run automatically in systems used for software development and testing without needing someone to watch it.

Use Cases

SajiCode can be used to build production software from start to finish. Its specialized agents can handle tasks for backend development, frontend design, quality assurance, security, and deployment. It can also analyze codebases in multiple programming languages to help agents understand the project better. For developers on the go, SajiCode can receive coding tasks through WhatsApp, and it can even act as a personal bot that replies to contacts in a learned chat style. It is also useful for automated tasks in CI/CD pipelines, such as running tests or reviewing code.

Pricing

Pricing information is not available in the provided article.

Vibes

Information about public reception or reviews is not available in the provided article.

Additional Information

SajiCode supports various AI language model providers, including local options like Ollama, as well as cloud-based services from OpenAI, Google, and Anthropic. The project is open source and welcomes contributions from the community for new skills, AI model support, and agent improvements. Its code is organized into clear directories for agents, tools, and other components, making it easier to understand and contribute to.