London AI startup Lua wants to give every business its own AI workforce with $5.8M seed round

AI agents are finally moving from demos to the day-to-day stack of real companies—and a London startup wants to make them feel less like science projects and more like hiring a new team. Lua, an operating system for humans and AI agents working side by side, has raised a $5.8 million seed round to help any business, technical or not, stand up its own AI “workforce” in hours instead of months.

The company is betting that the next wave of enterprise AI won’t be about one chatbot per workflow, but about orchestrating fleets of specialized agents that coordinate with employees, tools, and data sources like a real organization. For founders and investors, the question is no longer whether agents will matter, but who will own the layer that makes them dependable, governable, and ROI-positive inside existing businesses.

Who is backing Lua and why now

Lua’s $5.8 million seed round is led by Norrsken22, an Africa-focused tech growth fund, with participation from Flourish Ventures, Y Combinator, Phosphor Capital, 20VC, and other backers. Angels include founders and operators from companies such as Privy, Opendoor, and Nuiee Travel, signaling that the cap table blends AI-native investors with operators who have wrestled with automation in production environments.

The fresh capital will go toward expanding Lua’s developer community and a global network of implementation partners who can bring its platform into mid-market and enterprise accounts. That partner-centric approach suggests Lua is not only targeting early adopters, but also systems integrators and boutique consultancies that are looking to standardize how they build agentic workflows for clients.

From single agents to AI “workforces”

Many companies’ first contact with AI agents is a single assistant—say, for support triage or lead qualification—bolted onto existing tools. Lua’s core thesis is that this one-off model leaves too much value on the table and creates a maintenance problem: each agent is a snowflake with its own stack, prompts, and integrations.

Instead, Lua frames agents as a workforce that should be hired, onboarded, and managed collectively. Its platform lets teams design networks of agents that collaborate with humans and with each other, handle handoffs, and plug into existing systems across channels. In practice, that means a customer success agent can escalate to a billing agent, loop in a human account manager, and update a CRM, all under a shared orchestration layer.

Lua also positions itself as an “HR system for non-human employees,” emphasizing ownership and governance. Instead of renting a black-box assistant that lives on someone else’s infrastructure and data rails, customers are meant to own their agents and the data they generate, potentially compounding efficiency over time rather than just compounding usage costs.

What the platform actually does

Under the hood, Lua describes itself as an opinionated, full-stack agent platform. It bundles infrastructure, model orchestration, data pipelines, channel integrations, and monitoring into a single environment, leaving users to supply business logic and pick the systems their agents should connect to.

Developers can work through a command-line interface when they need granular control over agent behavior and infrastructure. Non-technical teams, meanwhile, interact with the same agents through a natural language interface, editing workflows and policies without leaving their browser. Both personas operate on the same artifacts in the same environment, which is meant to reduce the translation cost between operators and engineers.

The company claims that teams can stand up a functioning AI workforce—agents coordinating with humans and tools, integrated into existing systems—within hours of onboarding. That “hours, not months” promise is increasingly becoming table stakes in the agent platform space, but Lua backs it with usage metrics: since launching its developer platform in October 2025, the startup says revenue has been growing nearly 30% week-over-week. In February 2026 alone, more agents were created on Lua than during the entire period from launch through January.