Show HN: Praxos – AI 智能体的情境管理
9 分•作者: mogusian•6 个月前
大家好，我是 Lucas 和 Soheil，Praxos（<https://praxos.ai）的创始人。Praxos 是一个面向 AI 智能体的上下文管理器，提供构建可在生产环境中稳定运行的、有状态智能体所需的一切。
Praxos 可以解析任何数据源，从非结构化的 PDF 和 API 流，到对话消息，再到结构化数据库，并将它们转化为一个统一的知识图谱。图谱中的所有内容都经过语义类型标注，其关系也得到明确，从而将数据转化为一个清晰、可查询的理解宇宙，AI 可以无误地使用它。
无论您是需要查询问题的答案，还是需要以适合当前用例的方式提取数据，Praxos 都能做到，无需重复查询。这使得 AI 应用能够端到端地解析数据，然后根据数据采取行动，在单链和多链推理步骤中提供输出。中间、最终和用户编辑的输出可以添加回知识图谱，使 Praxos 能够动态学习。
在构建保险应用时，我们经常遇到部署 AI 的两个主要问题：
首先，大型语言模型（LLM）无法解析诸如财产明细表和保险单等文件。例如，财产明细表可能是一份包含 50 页的 Word、Excel 和 PDF 文档集合，详细说明了关于一系列物理财产的建造、使用和地理信息。要重建一个对象（财产），意味着要梳理这些文件，以建立数据之间的语义、概念、空间，有时甚至是隐含的联系。
结果：关系信息会丢失、留空或被虚构。
其次，重复调用搜索、检索和更新信息有时会导致级联错误。在诸如阅读文档、获取之前的用户信息、执行计算、存储它，然后将其呈现给用户的复杂任务中，这种情况变得更加频繁。
我们意识到，为了让 AI 提供更有用、更准确的响应，并正确使用文档中的关系，这些关系需要被明确。许多上下文信息没有使用文字来表示。反过来，这意味着我们无法通过编程直接与它们交互，而 LLM 每次都必须自己解释它们。这就是我们开始构建 Praxos 的时候。
我们为爱好者和早期使用者设置了一个自助选项，提供免费套餐（数据上限）。为了方便理解（并非双关语），这应该能满足您最多 200 页文档的需求。您可以在这里注册：<https://data.praxos.ai/sign-up>。
我们的第一个版本是一个 SDK，旨在满足您在所有数据提取、检索和更新方面的需求。
其工作原理如下：
组织信息：Praxos 将信息整理到本体中，本体是用于存储数据的结构化模式。这些允许您引入预定义的类型、属性和关系，从而指导知识图谱的构建和解释。
处理输入数据：Praxos 可以处理任何数据源，从 PDF 到表格数据、JSON 和类似对话的交流。提取是端到端执行的。您不需要 OCR、分块或预处理您的输入。处理就像传入您的文件并选择一个本体一样简单。
检索信息/记忆：对于每个查询，Praxos 通过结合图遍历技术、向量相似度和键值查找来搜索和检索相关的存储信息。搜索对象将返回实体/它们的关系，以及一个句子。
我们很乐意听取您的意见！请随意深入了解，并通过 Discord（<https://discord.gg/wVmrXD2dJA）与我们分享任何想法或建议。
您的反馈将帮助我们确定 Praxos 的未来发展方向！
查看原文
Hey HN! We're Lucas and Soheil, the founders of Praxos (<a href="https://praxos.ai">https://praxos.ai</a>). Praxos is a context manager for AI Agents, providing everything you need to build stateful agents that don't break in production.<p>Praxos can parse any data source, from unstructured PDFs and API streams to conversational messages, to structured databases, and transform them into a single Knowledge Graph. Everything in this graph is semantically typed and its relationships are made explicit, turning data into a clean, queryable universe of understanding that AI can use without making mistakes.<p>Whether you need to query for the answer to a question or to extract data in a way that makes sense for the current use case, Praxos does it all, with no requerying needed. This enables AI apps to parse data end-to-end, and then act on it to deliver outputs across single-chain and multi-chain reasoning steps. Intermediate, final, and user-edited outputs can be added back to the knowledge graph, allowing Praxos to learn on the fly.<p>When we were building in insurance, we often ran into two major problems deploying AI:<p>First, LLMs would prove incapable of parsing documents such as property schedules and insurance policies. For reference, a property schedule may be a 50-page collection of Word, Excel, and PDF documents detailing construction, usage, and geographical information about a collection of physical properties. Recreating one object (a property) would mean combing through the files establish semantic, conceptual, spatial, and sometimes implicit linkages between the data.<p>The outcome: relationship information would be lost, left blank, or hallucinated.<p>Second, repeated calls to search, retrieve, and update information would sometimes lead to cascading errors. This became more frequent across complex tasks such as reading a document, fetching previous user information, performing a calculation, storing it, and then presenting it to the user.<p>We realized that for AI to deliver more useful and accurate responses that correctly use relationships in the document, these relationships need to be made explicit. Much of the contextual information is represented without the usage of words. In turn, this means that we cannot directly interact with them programmatically, and LLMs are forced to interpret them themselves, every single time. That’s when we started building Praxos.<p>We've set up a self-serve option with a free tier (up to a data cap) for hobbyists and early-adopters. For context (no pun intended), this should cover you for up to 200 document pages. You can register here: <a href="https://data.praxos.ai/sign-up">https://data.praxos.ai/sign-up</a>.<p>Our first version is an SDK meant to cover you across all your data extraction, retrieval, and update needs.<p>Here's how it works:<p>Organizing information: Praxos sorts information into ontologies, which are structured schemas for storing data. These allow you to introduce predefined types, attributes, and relationships that guide how the knowledge graph is built and interpreted.<p>Processing input data: Praxos can handle any data source, ranging from PDFs to tabular data, JSONs, and dialog-like exchanges. Extraction is performed end-to-end. You don't need to OCR, chunk, or pre-process your inputs. Processing is as simple as passing in your file and selecting an ontology.<p>Retrieving information / memories: For each query, Praxos searches and retrieves related stored information by leveraging a combination of graph traversal techniques, vector similarity and key-value lookups. Search objects will return both the entities/their connections, as well as a sentence.<p>We’d love to hear what you think! Please feel free to dive in, and share any thoughts or suggestions with us over Discord (<a href="https://discord.gg/wVmrXD2dJA" rel="nofollow">https://discord.gg/wVmrXD2dJA</a>).<p>Your feedback will help shape where we take Praxos from here!