Jac

Jac is a programming language that tries to collapse the modern development stack into one coherent thing. Instead of writing Python for the backend, JavaScript for the frontend, and YAML for deployment, you write Jac and the compiler handles the rest. It outputs Python bytecode, JavaScript, or native machine code depending on what you're targeting, and every library from PyPI, npm, and native C ecosystems works directly without interop wrappers.

The language looks like Python and feels familiar if you've written Python, but it adds three ideas that change how you structure programs. First, meaning types turn large language models into language constructs. Instead of writing prompts manually, you annotate code with semantic meaning and the compiler generates optimized prompts automatically. That moves AI from an external API you call into something the language understands natively. Second, object spatial programming introduces nodes, edges, and walkers as core constructs. You model data as a graph and write computation that traverses it, which fits naturally for applications where relationships matter as much as records. Third, scale invariance means the same code runs unchanged in a terminal or across a distributed cluster. Deployment happens with a flag rather than a rewrite.

The practical effect is less glue code. Serializers, API routes, database bindings, and deployment manifests that normally require separate files and frameworks get generated by the compiler. You define a node, give it fields, and the runtime handles persistence. You start a service, and it provisions Kubernetes infrastructure automatically. The tradeoff is that you're committing to a new language and its opinions, but for projects that would otherwise span multiple languages and a lot of boilerplate, the reduction in surface area is significant.

Installation runs through a shell script that downloads a self contained binary for macOS or Linux. Windows support requires WSL for now. The project is MIT licensed and maintained by Jaseci Labs with an active community and regular releases. Documentation lives at docs.jaseci.org and covers the language tour, quick reference, and integration guides.

Jac targets developers building AI native applications who want to ship faster by working in one language end to end. If you're tired of context switching between frontend, backend, infrastructure, and prompt engineering, and you're willing to learn a new syntax to consolidate all of that, Jac offers a genuinely different approach to full stack development.