Today's AI Frameworks & SDKs: Fastest-Growing Projects — April 15, 2026
Today's AI Frameworks & SDKs, we're seeing a surge in tools that enable developers to build and manage complex AI systems. From MCP servers to multi-agent orchestration platforms, these tools are making it easier for developers to create sophisticated AI applications. With growth scores soaring across the board, it's clear that this space is heating up.
rohitg00/ai-engineering-from-scratch takes the top spot with a staggering growth score of 92.39 and over 2,810 stars. This repository provides a comprehensive guide for learning, building, and shipping AI systems, making it an invaluable resource for developers looking to get started in the field. Its massive popularity can be attributed to its hands-on approach, which resonates with developers eager to gain practical experience.
jxnxts/mcp-brasil comes in second with a growth score of 62.40 and 1,365 stars. This MCP server provides access to 41 public APIs in Brazil, making it an essential tool for developers working on AI projects that require data from the region. Its popularity stems from its ability to bridge the gap between AI models and real-world data.
yogthos/chiasmus has a growth score of 38.31 and 58 stars, but don't let its relatively low star count fool you – this repository is gaining traction quickly. Chiasmus is an MCP server that allows language models to access formal verification, providing a critical layer of security for AI applications. As developers become increasingly concerned with the reliability of their AI systems, tools like Chiasmus are likely to see significant growth.
Yuan-lab-LLM/ClawManager boasts a growth score of 26.31 and 561 stars, thanks to its innovative approach to managing AI agent instances. This Kubernetes-native control plane provides governed AI access, runtime orchestration, and reusable resources across multiple agent runtimes. As the need for efficient AI deployment grows, ClawManager is well-positioned to capitalize on this trend.
codefromkarl/ContextAtlas has a growth score of 26.12 and 24 stars, but its unique approach to context infrastructure for AI coding agents has piqued the interest of developers. This repository provides hybrid retrieval, project memory, and retrieval observability via CLI, MCP server, or embeddable library, making it an attractive solution for those looking to streamline their AI development workflow.
rcortx/kiwiq boasts a growth score of 25.36 and over 1,060 stars, despite having only three commits in the past month. This production-grade multi-agent orchestration platform has been battle-tested on over 200 enterprise AI agents, making it an attractive solution for developers looking for reliable and scalable tools.
OpenEnvision/Awesome-Multimodal-Modeling rounds out our list with a growth score of 22.87 and 230 stars. This repository provides an extensive collection of resources on multimodal modeling, covering MLLM, UMM, and NMM. As developers continue to explore the possibilities of multimodal AI applications, this repository is likely to remain a valuable resource.
rhino-acoustic/NeuronFS has a growth score of 20.20 and 136 stars, thanks to its innovative approach to OS-native constraint engines for LLM agents. This B-tree-based file system provides efficient token storage and retrieval, making it an attractive solution for developers looking to optimize their AI applications.
jnMetaCode/agency-orchestrator boasts a growth score of 19.38 and 285 stars, thanks to its innovative multi-agent framework that works with existing AI subscriptions. As developers look for ways to integrate multiple AI agents into their workflows, agency-orchestrator is well-positioned to capitalize on this trend.
JiaboLi-GitHub/renderdoc-mcp has a growth score of 18.10 and 109 stars, thanks to its innovative approach to analyzing GPU frame captures and debugging graphics pipelines using an MCP server. As AI applications become increasingly graphics-intensive, tools like renderdoc-mcp are likely to see significant growth.
rohitg00/ai-engineering-from-scratch takes the top spot with a staggering growth score of 92.39 and over 2,810 stars. This repository provides a comprehensive guide for learning, building, and shipping AI systems, making it an invaluable resource for developers looking to get started in the field. Its massive popularity can be attributed to its hands-on approach, which resonates with developers eager to gain practical experience.
jxnxts/mcp-brasil comes in second with a growth score of 62.40 and 1,365 stars. This MCP server provides access to 41 public APIs in Brazil, making it an essential tool for developers working on AI projects that require data from the region. Its popularity stems from its ability to bridge the gap between AI models and real-world data.
yogthos/chiasmus has a growth score of 38.31 and 58 stars, but don't let its relatively low star count fool you – this repository is gaining traction quickly. Chiasmus is an MCP server that allows language models to access formal verification, providing a critical layer of security for AI applications. As developers become increasingly concerned with the reliability of their AI systems, tools like Chiasmus are likely to see significant growth.
Yuan-lab-LLM/ClawManager boasts a growth score of 26.31 and 561 stars, thanks to its innovative approach to managing AI agent instances. This Kubernetes-native control plane provides governed AI access, runtime orchestration, and reusable resources across multiple agent runtimes. As the need for efficient AI deployment grows, ClawManager is well-positioned to capitalize on this trend.
codefromkarl/ContextAtlas has a growth score of 26.12 and 24 stars, but its unique approach to context infrastructure for AI coding agents has piqued the interest of developers. This repository provides hybrid retrieval, project memory, and retrieval observability via CLI, MCP server, or embeddable library, making it an attractive solution for those looking to streamline their AI development workflow.
rcortx/kiwiq boasts a growth score of 25.36 and over 1,060 stars, despite having only three commits in the past month. This production-grade multi-agent orchestration platform has been battle-tested on over 200 enterprise AI agents, making it an attractive solution for developers looking for reliable and scalable tools.
OpenEnvision/Awesome-Multimodal-Modeling rounds out our list with a growth score of 22.87 and 230 stars. This repository provides an extensive collection of resources on multimodal modeling, covering MLLM, UMM, and NMM. As developers continue to explore the possibilities of multimodal AI applications, this repository is likely to remain a valuable resource.
rhino-acoustic/NeuronFS has a growth score of 20.20 and 136 stars, thanks to its innovative approach to OS-native constraint engines for LLM agents. This B-tree-based file system provides efficient token storage and retrieval, making it an attractive solution for developers looking to optimize their AI applications.
jnMetaCode/agency-orchestrator boasts a growth score of 19.38 and 285 stars, thanks to its innovative multi-agent framework that works with existing AI subscriptions. As developers look for ways to integrate multiple AI agents into their workflows, agency-orchestrator is well-positioned to capitalize on this trend.
JiaboLi-GitHub/renderdoc-mcp has a growth score of 18.10 and 109 stars, thanks to its innovative approach to analyzing GPU frame captures and debugging graphics pipelines using an MCP server. As AI applications become increasingly graphics-intensive, tools like renderdoc-mcp are likely to see significant growth.