Islo Raises $15M to Scale Its Enterprise Software Factory
Islo raised $15 million to expand an AI-agent platform that automates engineering work through isolated, auditable Factory Lines, model testing and human approval gates.

Why it matters
- On November 17, 2026, Islo announced a $15 million investment led by Insight Partners and launched the Enterprise Software Factory.
- The funding supports product development, adoption and expansion of Islo as a dedicated product line within Incredibuild.
- Incredibuild acquired Kypso, founded by Islo’s leadership team, in 2025.
- Factory Lines combine AI-agent and script-based steps with configurable safety, security, performance and cost gates.
- An ARIMLABS pilot reached roughly 500,000 evaluation executions per day.
Islo announced a $15 million investment led by Insight Partners and launched its Enterprise Software Factory platform on November 17, 2026. The release puts AI agents into recurring engineering work while keeping security, verification and deployment under customer control, according to Islo.
Islo says the money will support product development, adoption and expansion of the company as a dedicated product line within Incredibuild. Incredibuild acquired Kypso, the startup founded by Islo’s leadership team, in 2025. The platform incorporates Incredibuild’s build, test and continuous-integration acceleration and connects that pipeline technology with controls for agent-generated work.
Islo frames the opportunity around a gap between code production and safe delivery. The company says higher agent output can increase the volume of code without ensuring that more software reaches production safely, which makes repeatable checks and approvals central to its enterprise pitch.
The launch covers more than code generation. Islo describes scheduled runs, automatic retries, isolated builds, cost records and approval gates across recurring engineering tasks, with conventional scripts retaining authority over fixed checks.
How do Factory Lines control AI-generated work?
Islo’s Factory Lines turn an engineering procedure into a repeatable sequence of agent and non-agent steps. Teams can assign implementation or review to an AI agent, then route builds, linters and tests through scripts so those checks do not depend on model judgment. The structure resembles an assembly line, with fixed checks placed around variable AI actions.
Runs can start when a team receives a new issue, a Slack message, a webhook or a timetable. A failed check sends the work back for another attempt. A risky action can stop the sequence and wait for human approval, according to Islo.
In Islo’s design, agents handle designated implementation or review steps, scripts retain authority over builds, linters and tests, and people retain authority over selected approvals. That division lets the platform automate a procedure without making the model responsible for every stage of verification.
What runs inside an isolated Machine?
Each Machine provides a persistent, hardware-isolated environment containing the repositories, services and dependencies required to build and run the customer’s software. Islo says this lets an agent verify a proposed change against the customer’s actual stack rather than a generic development environment.
Cloud hosting lets multiple Machines keep running without an engineer staying logged in. The configuration supports work triggered by an issue, message, webhook or timetable that must continue through build, test and retry cycles without continuous human supervision.
Islo’s security controls operate within that environment. Outbound connections remain denied unless the customer permits them. Gateway injection keeps raw credentials out of the sandbox, and the system records commands, file modifications and network activity.
Customers can also run Islo in their own cloud and set spending limits for each line, according to the product page. They can attach sessions, actions, outbound calls and usage costs to an auditable run record, linking agent activity with its cost and network behavior.
Early customer uses listed by Islo include refactoring, code review, quality assurance, bug fixing, issue triage and long-tail feature delivery. Teams can set gates for safety, testing, functionality, security, performance and cost, and can require human sign-off for selected actions.
Can more code be delivered safely?
“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. Islo’s stated response is to make the checks and approvals surrounding a change as repeatable as its generation.
Islo frames the governance stakes in concrete terms: customers will increase the volume and autonomy of agent-generated changes, but review and verification must not become the next constraint. Its design puts deterministic scripts and explicit approval gates inside the workflow instead of treating the model as the final authority.
Islo’s commercial proposition centers on control rather than generation alone. The company positions Factory Lines as a way to turn repeatable engineering procedures into governed workflows, with audit records for each run and explicit gates for cost, performance and security.
What changes when Databricks and model choice enter the workflow?
Databricks is slated to become Islo’s backend inference partner. Islo says the integration will give customers access to proprietary and open AI models and let joint customers run factories against their own data without giving up privacy or control.
Model execution and access to Databricks data will sit inside the Software Factory, in the same controlled environment as the engineering workflow. Islo says that placement brings model resources under the platform’s security, isolation and audit controls.
Teams can also use Claude Code, Codex or other agent tools and choose different models for individual stages, according to Islo. They can replace one component without rebuilding an entire Factory Line, a design choice Islo calls model independence.
How does Islo test a model change before deployment?
Islo says proposed changes to a model, prompt or agent harness can first run against historical tickets in the customer’s environment. The evaluation then examines the result, hidden tests, side effects, attempts and cost.
Under Islo’s evaluation setup, simulated, state-preserving versions of GitHub, Slack and Linear keep the agent from finding the eventual resolution before it attempts the ticket. The platform can then assess the attempt against the customer’s own historical work and acceptance conditions rather than a general vendor benchmark.
The evaluation covers software outcomes as well as operating cost. In addition to the result and hidden tests, Islo examines side effects and the number of attempts, then includes cost in the evidence for a proposed change.
Islo says the resulting record reflects the customer’s work rather than a general vendor benchmark. Buyers can use that customer-specific evidence to test whether a prompt, harness or model change alters outcomes on software they already know.
Which workloads are running now?
Islo reports that Rite uses the platform for feature delivery and bug fixing. The company also says Incredibuild expanded its own use from automated pull-request review to bug fixing, feature delivery and continuous-integration triage.
An ARIMLABS pilot grew to roughly 500,000 evaluation executions per day, according to Islo. The figure supplies a concrete measure of the recurring workload that Islo’s evaluation architecture can process.
Islo says those deployments point to an initial workload of continuous, recurring tasks whose inputs and acceptance checks can be specified in advance. The company identifies maintenance, testing and triage backlogs with objective checks as the immediate commercial opportunity.
The next test for Islo is whether its controls can keep review and verification from becoming the next constraint as customers increase the volume and autonomy of agent-generated changes; the $15 million investment gives the dedicated product line resources to test that proposition at greater scale.
Source: unite.ai
More from Rebecca Stone
Show full bio
Correspondent covering companies, deals and strategy in artificial intelligence and machine learning at AI In Context.
728 articles
Related articles
- Islo Raises $15 Million for Controlled AI Software Factories
- AgentKit, Expanded Evals, and RFT for Agents Ship Together
- AMD Says AI Agents Now Auto-Fix 75% of Radeon Software Bugs
- OpenAI to Acquire Promptfoo and Build Red-Teaming Into Frontier
- Deloitte Launches Edisyl to Sell Expert Knowledge as an AI Asset