It enables building AI systems that are modular, controllable and aligned with real-world requirements. Think of it as "LEGO blocks for AI agents"."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: A New Paradigm for Building Agentic AI Pipelines, Atomically"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["atomic","agent"],"value":"https://github.com/KennyVaneetvelde/atomic_agents"}},"_tags":["story","author_KennyVan","story_40959854","show_hn"],"author":"KennyVan","created_at":"2024-07-14T09:23:06Z","created_at_i":1720948986,"num_comments":0,"objectID":"40959854","points":1,"story_id":40959854,"story_text":"Atomic Agents is a Python framework that brings software engineering principles to AI agent development. Key features:\n- Focus on practical business applications rather than fancy autonomous clickbait demos\n- Modular design inspired by Atomic Design, breaking AI systems into reusable components\n- Structured input/output using Pydantic schemas for predictable behavior\n- Separation of concerns (prompts, tools, memory) for fine-grained control\n- Easy chaining of agents and tools to build complex workflows
It enables building AI systems that are modular, controllable and aligned with real-world requirements. Think of it as "LEGO blocks for AI agents".","title":"Show HN: A New Paradigm for Building Agentic AI Pipelines, Atomically","updated_at":"2024-09-20T17:28:03Z","url":"https://github.com/KennyVaneetvelde/atomic_agents"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"pancsta"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["atomic","agent"],"value":"Very informative wiki, thank you, I will definitely use it. So Ive made my own "AI Agents framework" [0] based on actor model, state machines and aspect oriented programming (released just yesterday, no HN post yet) and I really like points 5 and 7:
5: Unify execution state and business state\n 8. Own your control flow\nFeel free to reach out.
[0] https://github.com/pancsta/secai
[1] https://youtu.be/0VJzO1S-gV0"},"story_title":{"matchLevel":"none","matchedWords":[],"value":"12-factor Agents: Patterns of reliable LLM applications"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/humanlayer/12-factor-agents"}},"_tags":["comment","author_pancsta","story_43699271"],"author":"pancsta","children":[43706597,43714012,43714846],"comment_text":"Very informative wiki, thank you, I will definitely use it. So Ive made my own "AI Agents framework" [0] based on actor model, state machines and aspect oriented programming (released just yesterday, no HN post yet) and I really like points 5 and 7:
5: Unify execution state and business state\n 8. Own your control flow\nFeel free to reach out.
[0] https://github.com/pancsta/secai
[1] https://youtu.be/0VJzO1S-gV0","created_at":"2025-04-16T09:09:51Z","created_at_i":1744794591,"objectID":"43703196","parent_id":43699271,"story_id":43699271,"story_title":"12-factor Agents: Patterns of reliable LLM applications","story_url":"https://github.com/humanlayer/12-factor-agents","updated_at":"2025-06-26T12:31:28Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"DeadlyPretzel"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["atomic","agent"],"value":"For those who don't know yet, Atomic Agents ( https://github.com/KennyVaneetvelde/atomic_agents ) is designed to be modular, extensible, and easy to use. Components in the Atomic Agents Framework should always be as small and single-purpose as possible, similar to design system components in Atomic Design. Even though Atomic Design cannot be directly applied to AI agent architecture, a lot of ideas were taken from it. The resulting framework provides a set of tools and agents that can be combined to create powerful applications. The framework is built on top of Instructor and uses Pydantic for data validation and serialization.\nFor those who have been following it for a bit, it just got a lot easier to build new agents using any client supported by Instructor, including local agents.\nI highly recommend checking out:\n- The basic custom chatbot example: https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...
- Yelp agent to help find restaurants on yelp: https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...\nThis demo essentially shows how an agent in Atomic Agents can be given a schema and figure out the best way on its own to ask the user the right questions in order to gather the necessary information for performing the API call. This logic can essentially be applied to any filterable API or endpoint, ... such as for a webshop's products (hint hint, product idea)\n- Deep multi-agent research example (like perplexity): https://github.com/KennyVaneetvelde/atomic_agents/tree/main/...\n- Agent orchestration demo (in other words, letting an agent outsource tasks to other agents): https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...
- Easily sharing dynamic context between two atomic agents: https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...
More examples: https://github.com/KennyVaneetvelde/atomic_agents/tree/main/...\nDocs: https://github.com/KennyVaneetvelde/atomic_agents/tree/main/..."},"story_title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["atomic","agent"],"value":"Atomic Agents 0.1.44 Released"},"story_url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["atomic","agent"],"value":"https://github.com/KennyVaneetvelde/atomic_agents"}},"_tags":["comment","author_DeadlyPretzel","story_40753359"],"author":"DeadlyPretzel","comment_text":"For those who don't know yet, Atomic Agents ( https://github.com/KennyVaneetvelde/atomic_agents ) is designed to be modular, extensible, and easy to use. Components in the Atomic Agents Framework should always be as small and single-purpose as possible, similar to design system components in Atomic Design. Even though Atomic Design cannot be directly applied to AI agent architecture, a lot of ideas were taken from it. The resulting framework provides a set of tools and agents that can be combined to create powerful applications. The framework is built on top of Instructor and uses Pydantic for data validation and serialization.\nFor those who have been following it for a bit, it just got a lot easier to build new agents using any client supported by Instructor, including local agents.\nI highly recommend checking out:\n- The basic custom chatbot example: https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...
- Yelp agent to help find restaurants on yelp: https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...\nThis demo essentially shows how an agent in Atomic Agents can be given a schema and figure out the best way on its own to ask the user the right questions in order to gather the necessary information for performing the API call. This logic can essentially be applied to any filterable API or endpoint, ... such as for a webshop's products (hint hint, product idea)\n- Deep multi-agent research example (like perplexity): https://github.com/KennyVaneetvelde/atomic_agents/tree/main/...\n- Agent orchestration demo (in other words, letting an agent outsource tasks to other agents): https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...
- Easily sharing dynamic context between two atomic agents: https://github.com/KennyVaneetvelde/atomic_agents/blob/main/...
More examples: https://github.com/KennyVaneetvelde/atomic_agents/tree/main/...\nDocs: https://github.com/KennyVaneetvelde/atomic_agents/tree/main/...","created_at":"2024-06-21T20:17:11Z","created_at_i":1719001031,"objectID":"40753360","parent_id":40753359,"story_id":40753359,"story_title":"Atomic Agents 0.1.44 Released","story_url":"https://github.com/KennyVaneetvelde/atomic_agents","updated_at":"2024-09-20T17:17:09Z"}],"hitsPerPage":5,"nbHits":27,"nbPages":6,"page":0,"params":"query=atomic-agent&hitsPerPage=5&advancedSyntax=true&analyticsTags=backend","processingTimeMS":28,"processingTimingsMS":{"_request":{"roundTrip":18},"afterFetch":{"merge":{"mergeLoop":{"prepareNextHit":3,"total":3},"total":3},"total":3},"fetch":{"query":10,"scanning":12,"total":23},"total":28},"query":"atomic-agent","serverTimeMS":29}