供资
Shackleton AI 融资 500万种子将AI Agents投入整个建筑工程
在 Google 上的首选来源中添加 Unite.AI
Shackleton AI拥有由Battery Ventures牵头的500万美元的种子资金融资，因为它从隐形中脱颖而出，有一个用于开发和建筑管理的代理平台。 纽约公司的目标是在建筑项目的计划、财务、实地报告和通讯之间开展行政工作。
Brick & Mortar Ventures, Trust Ventures，关联Ventures，动机伙伴，Thayer投资伙伴，Act One Ventures，脑暴Ventures, Silence VC，以及其他投资者参加了会议。 沙克尔顿计划利用资本来扩展工作流程，其AI agents可以管理和加快其上市努力。
根据该公司的发布公告，2026年6月的平台发布，现在管理超过15个州超过10亿美元的项目价值。 该数字描述了通过软件管理的建筑项目，而不是沙克尔顿的收入或估价。 该公司的客户公司包括从事多家庭、工业和商业建设的中型总承包商，以及红龙虾、狗豪斯和胡椒午餐等多地点品牌。
内建总承包商
共同创始人迈克尔·拉格布，哈桑·亚辛，奥莱克桑德尔·卡尔特瑟夫在建造了以技术为动力的商业总承包商欧内斯特后开发了该平台，整个美国东南部的收入达到3500万美元。 他们的出发点是难以在互不连接的软件中协调实际项目。
称 Ragheb, Shackleton的共同创始人兼首席执行官在公告中表示，该公司辩称，项目经理在投标、时间安排、文件和财务工具之间损失了宝贵的时间来传递信息。
电池公司总合伙人Michael Brown指出，这种运营经验是投资理由的一部分。 “Shackleton团队建立这个平台是为了经营自己的业务
AI Agents 项目记录
Shackleton将其产品描述为一种代理操作系统，涵盖选址、施工前、项目和外地管理、财务、通信和项目周转。 在此，“操作系统”是指协调建筑工程的共享业务软件平台。
agents的任务包括：起草索取资料的请求，或索取索取资料书；对逾期未交文件采取后续行动；在付款请求得到批准之前查明预算差异；将字段指出转换为每日报告。 《索取资料书》记录了一个需要澄清项目要求的问题，同时提交文件材料、设备或其他拟议工作供审查。
该公司的网站描述了一个共同的项目模式，包括投标、预算、索取资料书、提交书、变更单和实地报告。 意图的优势是上下文：agent可以跨相关记录工作，而不是只看到一个应用程序内部的信息。 Shackleton介绍了具体角色的agents功能，包括外地工作、规划、发展、采购、法律工作、增长以及办公室和财务业务。
它的浏览器和外地接口也处理不同的工作条件。 办公室用户可以协调时间表、里程碑和关键路径；外地小组可以访问图纸并获取现场更新。 公司称床单仍可读且可通知离线，然后在连接返回时调节。 这些是沙克尔顿描述的产品能力，而不是独立测试的结果。
连接项目活动到会计
Shackleton的产品概览将项目管理、管道和建议书、投标管理、外地报告、财务和通信中心结合起来。 它描述了边标比较和一个分包商门户，项目通信与相关工作相连。 通信枢纽列出Google工作空间和Microsoft365集成。
该公司的整合目录分别列出QuickBooks Online,QuickBooks桌面，Sage 300建筑与房地产，以及Sage Intact. 视连接器而定，所绘制的记录包括项目、供应商或分包商、成本代码、承付款、发票、账单和预算。 这使该平台能够将工作活动与现有的会计系统联系起来，而不是要求每一份财务记录都通过手工重新制作。
对于处理预算例外的agent来说，这些连接是必然的。 有必要将项目变化与相关成本记录和核准工作流程联系起来。 编制一份可信的摘要只是完成这项行政任务的一部分。
人类认可仍然是工作流程的一部分
发布公告称 agents运行于整个平台，同时人们通过审查和批准工作流程保留监督权。 Shackleton的长期雄心是建设项目管理大部分自己的协调，但雄心应该与发布描述的行政能力区分开来。
Ragheb 称多地点客户提前完成了35%的建设，预算也达到了18%。 这些是公司报告的结果，不是每次部署的保证，也不是对AI agents贡献的独立衡量。
资金紧随更广泛的推动，将AI引入建筑工作流程，包括Trunk Tools等平台及其以建筑为重点的agents. Shackleton的办法是巩固基本项目记录，下一阶段将扩大工作流程范围，同时使项目决定和财务核准对负责交付的小组负责。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Funding
Shackleton AI Raises $5 Million Seed to Put AI Agents to Work Across Construction
Add Unite.AI to your preferred sources on Google
Shackleton AI has raised $5 million in seed funding led by Battery Ventures as it emerges from stealth with an agentic platform for development and construction management. The New York company is targeting the administrative work that sits between a construction project’s plans, finances, field reports, and communications.
Brick & Mortar Ventures, Trust Ventures, Correlation Ventures, Motivate Partners, Thayer Investment Partners, Act One Ventures, Brainstorm Ventures, Silence VC, and other investors participated. Shackleton plans to use the capital to expand the workflows its AI agents can manage and accelerate its go-to-market efforts.
According to the company’s launch announcement, the platform launched in June 2026 and now manages more than $1 billion in project value across more than 15 states. That figure describes construction projects managed through the software, rather than Shackleton’s revenue or valuation. Its customers include midsize general contractors working in multifamily, industrial, and commercial construction, alongside multi-location brands such as Red Lobster, Dog Haus, and Pepper Lunch.
Built Inside a General Contractor
Co-founders Michael Ragheb, Hassan Yasin, and Olexandr Kaltsev developed the platform after building Ernest, a technology-enabled commercial general contractor, to $35 million in revenue across the Southeastern United States. Their starting point was the difficulty of coordinating real projects across disconnected software.
“I was spending more time moving data between systems than running my jobs at Ernest,” said Ragheb, Shackleton’s co-founder and CEO, in the announcement. The company argues that project managers lose valuable time transferring information between bidding, scheduling, document, and financial tools.
Battery Ventures general partner Michael Brown pointed to that operating experience as part of the investment rationale. “The Shackleton team built this platform to run their own business before bringing it to market,” he said.
AI Agents Working Across Project Records
Shackleton describes its product as an agentic operating system spanning site selection, preconstruction, project and field management, finance, communications, and project turnover. Here, “operating system” refers to a shared business software platform coordinating construction work.
The agents’ tasks include drafting requests for information, or RFIs; following up on overdue submittals; identifying budget variances before payment requests reach approval; and converting field notes into daily reports. An RFI records a question requiring clarification about project requirements, while submittals commonly document materials, equipment, or other proposed work for review.
The company’s website describes a common project model containing bids, budgets, RFIs, submittals, change orders, and field reports. The intended advantage is context: an agent can work across related records instead of seeing only the information inside a single application. Shackleton presents role-specific agents for functions including field work, planning, development, procurement, legal work, growth, and office and finance operations.
Its browser and field interfaces also address different working conditions. Office users can coordinate schedules, milestones, and critical paths; field teams can access drawings and capture updates on site. The company says sheets remain readable and annotatable offline, then reconcile when connectivity returns. These are product capabilities described by Shackleton, rather than results from independent testing.
Connecting Project Activity to Accounting
Shackleton’s product overview combines project management, pipeline and proposals, bid management, field reporting, finance, and a communication hub. It describes side-by-side bid comparison and a subcontractor portal, with project communications linked to the relevant job. The communication hub lists Google Workspace and Microsoft 365 integrations.
The company’s integration directory separately lists QuickBooks Online, QuickBooks Desktop, Sage 300 Construction and Real Estate, and Sage Intacct. Depending on the connector, the mapped records include projects, vendors or subcontractors, cost codes, commitments, invoices, bills, and budgets. This gives the platform a way to connect job activity to an existing accounting system rather than requiring every financial record to be recreated by hand.
For an agent handling a budget exception, those connections are consequential. A useful alert needs to relate a change in the project to the relevant cost record and approval workflow. Generating a plausible summary is only one part of completing that administrative task.
Human Approval Remains Part of the Workflow
The launch announcement says agents operate across the platform while people retain oversight through review and approval workflows. Shackleton’s longer-term ambition is for construction projects to manage much of their own coordination, but that ambition should be distinguished from the administrative capabilities described at launch.
Ragheb says multi-location customers have completed buildouts up to 35% ahead of schedule and as much as 18% under budget. Those are company-reported outcomes, not a guarantee for every deployment or an independently established measure of the AI agents’ contribution.
The funding follows a broader push to bring AI into construction workflows, including platforms such as Trunk Tools and its construction-focused agents. Shackleton’s approach centers on consolidating the underlying project record and embedding assistance across it. The next stage will be expanding that workflow coverage while keeping project decisions and financial approvals accountable to the teams responsible for delivery.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.