AI 热点

Islo 获 Insight Partners 领投 1500 万美元,打造可控企业 AI 软件工厂

Unite.AI更新于 11月18日 00:14

Islo Secures $15 Million to Build Controlled AI Software Factories

AI 导读

Islo 获得 Insight Partners 领投的 1500 万美元投资,用于扩展其 Enterprise Software Factory 平台,把智能体执行环境、访问控制和持续验证结合起来,让团队在保留生产变更审批权的同时自动化重复开发工作。平台以 Islo Machines 微虚拟机提供隔离开发环境,并通过网关管控智能体对外部系统的访问。Islo 同时宣布与 Databricks 合作提供模型推理并连接企业数据。

正文

Islo has secured a $15 million investment led by Insight Partners to expand its Enterprise Software Factory platform, betting that the next bottleneck in AI-assisted development is everything that happens after an agent starts writing code.

Part of the Incredibuild family of products, Islo combines agent execution environments, access controls, and continuous verification so engineering teams can automate repeatable work while retaining approval over production changes. The investment will support product development and adoption of the platform as a dedicated product line.

The announcement also outlines a partnership with Databricks to provide model inference and connect agent workflows to enterprise data. Together, the moves target a practical enterprise problem: producing more code is useful only if teams can reliably test, secure, review, and deliver it.

Islo describes its approach as a software factory. Rather than asking a developer to initiate every task in a chat window, teams define workflows that respond to events or run on a schedule. The company says early customers are applying that approach to refactoring, code review, quality assurance, bug fixes, issue triage, and smaller feature requests.

Each workflow has gates for testing, functionality, security, performance, and cost. Engineering teams decide where human approval is required, with sign-off retained before AI-generated changes reach production.

“AI can dramatically increase the amount of code an engineering organization produces, but that doesn’t automatically translate into more software being delivered safely,” said Adam Gold, Islo’s chief product officer, in the announcement.

That distinction changes what an agent platform must provide. Generating a patch is one step; running the application, checking the patch against its dependencies, and showing a reviewer what happened are separate engineering tasks.

Why Persistent Development Environments Matter

The foundation is Islo Machines, isolated development environments that can run an application’s services alongside an agent. The company describes these as microVMs, lightweight virtual machines with their own operating-system kernels, rather than containers sharing the host kernel.

A Machine can include dependencies, configuration, a browser, databases such as Postgres, and services such as Redis. This gives an agent a working application to exercise instead of asking it to reason about source files alone.

Persistence addresses another limitation of short-lived sandboxes. Snapshots preserve installed dependencies and database state, letting subsequent work resume from a prepared environment. Jobs can continue after a developer disconnects, which is important for tasks that take hours or span multiple iterations.

Islo also offers deployment inside a customer’s virtual private cloud. For enterprises, that makes the location of the runtime part of the deployment decision alongside the choice of model and agent tooling.

An isolated computer still needs controlled access to repositories, issue trackers, and other systems. Islo’s gateway handles that boundary by checking outbound requests and supplying credentials when an approved request leaves the environment.

In the company’s described workflow, an agent uses a placeholder binding rather than receiving the underlying secret. Gateway rules can restrict destinations, URL paths, and request methods. An optional model-based judge can assess request intent alongside those static rules.

This separates permission to perform a task from possession of a reusable API key. An agent reviewing a pull request, for example, can receive access appropriate to that workflow without being handed the same broad credentials used by a developer.

Allowed and denied requests are logged, with export to security monitoring systems. These records give teams a way to reconstruct which external services an agent contacted and which operations its policies blocked. The effectiveness of that boundary still depends on the permissions and rules an enterprise configures.

The announced Databricks relationship adds another layer: inference, the process of running a model to generate a response, and access to the enterprise data needed for agent-driven work.

Islo’s Databricks integration page describes connecting an existing workspace through brokered credentials. Agents can use models available through Databricks Model Serving and request scoped access to the customer’s compute and data.

Data processing remains on the customer’s Databricks clusters, while Islo supplies the isolated agent environment and controlled connection. The integration records requests and jobs for review, supporting workflows such as data engineering, analysis, and batch processing.

The release describes inference as the starting point for a broader product integration and Islo’s inclusion in Databricks’ Built on Program. This connects the software-factory strategy to an existing enterprise data platform, rather than requiring teams to duplicate their data in a separate agent tool.

Building on Incredibuild’s Delivery Infrastructure

Incredibuild brings distributed execution and caching technology to the platform. Distributed execution spreads suitable build or test work across compute resources; caching reuses previous outputs when the relevant inputs have not changed. Both can shorten the feedback cycle between an agent proposing a change and a team learning whether it works.

The relationship follows Incredibuild’s acquisition of Kypso in 2025. Kypso’s engineering agents and workflow automation contributed to the company’s expansion beyond build acceleration into a broader software delivery platform.

For Islo, the $15 million investment backs the infrastructure around autonomous engineering: environments that remain available, permissions that can be enforced, and checks that connect generated code to release decisions. Its opportunity is to make repeated development work dependable enough to delegate, while keeping engineers responsible for what ultimately ships.

来源

Unite.AI