Islo Raises $15M from Insight Partners for AI-Agent 'Software Factory'
Everyone can now write code with AI, but getting it safely into production is still hard. Islo, part of the Incredibuild family, raised $15M led by Insight Partners for a "software factory": a set of automated pipelines where AI agents handle refactoring, review, bug fixes and issue triage. For teams, this is a signal that checks and controls around AI-written code are becoming a product category of their own.
AI-processed from TNW; edited by Hamidun News
Assistants like Copilot and Cursor have made code cheap: it can now be generated dozens of times faster. But for large companies, this has created a new problem. All that code still has to be reviewed, tested, checked for vulnerabilities and rolled out carefully. Israeli startup Islo, part of the Incredibuild product family, has decided to tackle exactly this and raised $15M. The round was led by Insight Partners.
A caveat is in order: the source material from TNW is marked as contributed, meaning it is essentially a company press release. It contains no independent data on customer results, only claims made by the participants themselves.
What a 'software factory' is and how it works
Islo calls its platform the Enterprise Software Factory. The idea is simple if you think of an ordinary factory: there are assembly lines, and each line performs one repetitive operation. In Islo's terms, these are factory lines, meaning automated workflows in which AI agents do the work.
According to the company, early customers already run such lines for several tasks:
- code refactoring;
- change review;
- quality assurance (QA);
- bug fixing;
- initial triage of incoming issues (issue triage);
- implementing rare "long-tail" features that people never get around to.
The lines can run continuously. They are triggered either by an event (for example, a new ticket appears) or by a schedule set by the team. Each line has its own requirements for security, tests, performance and cost. At the relevant stages, human approval can be made mandatory, so teams can decide where control is actually needed.
For example: a repository has accumulated hundreds of small bugs that nobody picks up. An agent takes a ticket, reproduces the error in an isolated environment, proposes a fix, and runs the tests and build. After that, the developer just clicks "accept" or rejects the change.
Databricks as the backend and the 'bring your own' principle
Alongside the funding news, Islo announced an upcoming partnership with Databricks. The company will become the inference provider, meaning the service that runs the models, for the platform. Customers will have access to both proprietary and open models. In addition, the company says it can work directly with customer data in the lakehouse without compromising privacy.
The flexibility here is notable: Islo does not impose a technology stack. The organization itself chooses the language models, agents, agent frameworks and compute infrastructure. For enterprise buyers this matters: many already have contracts with specific model providers and strict requirements about where their data resides.
Why investors are betting on control rather than speed
Behind Islo stands the experience of Incredibuild, which supplies development infrastructure to companies such as Citibank, Bosch and Adobe. That is where the platform's built-in technologies for accelerating tests, builds and continuous integration (CI) come from.
Islo CEO Shimon Hason puts the thesis this way: the speed of writing code is no longer the bottleneck. What slows things down are the processes around it: isolated environments, constant verification and the "gates" before a release. Insight Partners Vice President David Stark makes the same point from the investor's side: agents need to respect data access rights, changes need to be verified, and corporate policies must not be violated.
What this means for practitioners
For developers and tech leads, the takeaway is this: the next stage of AI adoption in development is not another autocomplete, but automation of repetitive processes with boundaries and audit trails. If you are building internal agents, it is worth planning from the start for isolated execution environments, verification of results through tests, and points of manual approval.
There are also open questions. The company has not disclosed valuation, quality metrics (how many of the agents' changes people accept) or pricing. The market is crowded too: both large players and other startups focused on autonomous coding agents are working on similar things. So the $15M is more a bet on a hypothesis that, in enterprise AI coding, trust will be the core product rather than generation speed. The first large production deployments will test that.
- Nov 21, 2026Microsoft outage: nearly 83,000 Outlook complaints on Downdetector, Teams and Copilot down too
- Nov 21, 2026MLPerf Training Adds Nvidia Rubin and TPU Ironwood in First Appearance: 153 Systems
- Nov 21, 2026Microsoft Turns Copilot Into an "Operating System for Work": Word and Excel Become Agent Plugins
- Nov 21, 2026Google to split Gemini chat history from model training, so you can opt out without losing your history
Key Questions Answered
What a 'software factory' is and how it works?
Islo calls its platform the Enterprise Software Factory. The idea is simple if you think of an ordinary factory: there are assembly lines, and each line performs one repetitive operation. In Islo's terms, these are factory lines, meaning automated workflows in which AI agents do the work.
What are the key details on Databricks as the backend and the 'bring your own' principle?
Alongside the funding news, Islo announced an upcoming partnership with Databricks. The company will become the inference provider, meaning the service that runs the models, for the platform. Customers will have access to both proprietary and open models. In addition, the company says it can work directly with customer data in the lakehouse without compromising privacy. The flexibility here is notable: Islo does not impose a technology stack.
Why investors are betting on control rather than speed?
Behind Islo stands the experience of Incredibuild, which supplies development infrastructure to companies such as Citibank, Bosch and Adobe. That is where the platform's built-in technologies for accelerating tests, builds and continuous integration (CI) come from. Islo CEO Shimon Hason puts the thesis this way: the speed of writing code is no longer the bottleneck. What slows things down are the processes around it: isolated environments, constant verification and the "gates" before a release.
Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews
Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.