Leah Launches Harness Builder for Enterprise AI
The Maestro platform capability lets business users describe objectives in plain language and have AI construct connected applications.
Updated on Nov. 5, 2026 in Artificial Intelligence
Live Poll
Do you think AI running more business work will make everyday services better for you?
London-based enterprise agentic AI company Leah has launched Harness Builder, a capability within its Maestro platform. It allows business users to describe objectives in ordinary language and have AI build applications around them.
Why it matters
The tool lets non-technical staff direct AI-built applications while connecting to existing enterprise systems, policies and human oversight. It reflects a broader push to make agentic AI usable beyond engineering teams.
The figures come from company announcement data: Leah serves more than 400 global enterprises and was founded in London in 2015. Harness Builder's measured performance and customer results were not disclosed.
The players
Leah
Leah is a London-based enterprise agentic AI company formerly known as ContractPodAi, backed by SoftBank and Insight Partners and serving clients including Microsoft, Google, OpenAI and Anthropic.
Sarvarth Misra
Sarvarth Misra is the co-founder and CEO of Leah, leading the company's enterprise agentic AI strategy.
The details
Harness Builder identifies the capabilities a task requires and creates a connected harness, such as one that reviews invoices against billing guidelines. Maestro determines how specialist agents, business knowledge, tools and enterprise systems work together.
Timeline
- January 2026: ContractPodAi rebranded as Leah.
- November 5, 2026: Leah announced the launch of Harness Builder.
Roadmap
Harness Builder extends Leah's Maestro platform, which coordinates specialist agents and enterprise systems, into natural-language application construction. The launch follows the company's rebrand from ContractPodAi in January 2026 as it moves from contract management toward general enterprise agents.
Business teams may be able to describe workflows in plain language instead of waiting on engineering resources to build them. Organizations adopting such tools will still need to weigh governance, oversight and integration with existing systems.
The takeaway
Teams evaluating agentic AI should check how any tool connects to existing systems and what human oversight it supports before relying on AI-built applications. Early adoption works best with clear policies defined up front.
Further reading
Explore more in Artificial Intelligence coverage.
Source note: This article includes information reported by LawFuel.
Live Poll
Do you think AI running more business work will make everyday services better for you?
