Jaseci provides the Jac programming language, an AI‑native full‑stack language that lets developers write backend logic, React‑style UI components, and large‑language‑model integrations in a single .jac file. Jac compiles to native Python, JavaScript, and machine code, giving seamless access to PyPI, npm, and system libraries while handling state management, type safety, and automatic endpoint generation. The Jaseci stack also includes plugins for client‑side React generation and one‑command cloud deployment, simplifying development and scaling of AI‑enhanced web applications.
Funding
Funding not disclosed
Founders
Product
Problem
Developers building AI‑enhanced applications must juggle multiple languages and toolchains—Python for backend logic, JavaScript/TypeScript for frontend UI, and separate build pipelines for AI model integration—leading to context switching, duplicated code, and complex deployment workflows.
Solution
Jaseci offers Jac, an AI‑native full‑stack programming language that unifies backend, frontend, and AI code in a single syntax. Jac compiles to native Python bytecode, JavaScript (React), and machine code, giving developers seamless access to the PyPI, npm, and native ecosystems without writing separate codebases. The Jaseci stack includes plugins for client‑side React component generation (jac‑client) and automatic REST endpoint creation, scaling, and containerization (jac‑scale), enabling one‑command deployment from a laptop to cloud environments. By treating large language models as first‑class citizens and providing graph‑based object‑spatial programming, Jac lets developers embed LLM calls directly in their code while preserving type safety and human‑readable structure.
Target Audience
Primary users are full‑stack developers and AI engineers who need to build web applications, agentic systems, or knowledge‑graph services with integrated large language model capabilities, as well as teams seeking rapid prototyping and cloud‑native deployment.
Features
- Single `.jac` file can contain backend logic, React‑style UI components, and AI integrations, eliminating language fragmentation
- Multi‑target compilation produces pure Python, pure JavaScript, and native machine code, granting full access to PyPI, npm, and system libraries
- jac‑client plugin auto‑generates React components, handles npm dependency management, and provides hot‑module replacement for rapid UI development
- jac‑scale plugin turns public walkers into FastAPI endpoints, adds Swagger docs, SQLite/Mongo persistence, and supports Docker/Kubernetes deployment with a single CLI flag
- Built‑in LLM primitives (`by llm()`) enable direct, type‑safe calls to large language models from any part of the application
- Object‑Spatial Programming model supports graph‑structured domains and mobile walkers for agentic AI systems
- One‑command installation and project scaffolding (`jac create --use client`) streamline setup and local development