In most organisations today, software still follows a structure where work is broken into steps, passed between systems, and overseen by people at every stage. Kenyan startup, Lua believes that model is reaching its limits.

The Nairobi-based company has raised $5.8 million (approximately KES 748 million) in seed funding to expand a platform designed to build AI “agents”, software systems capable of executing entire business processes from start to finish.

The round was led by Norrsken22, with participation from Y Combinator, Flourish Ventures, P1 Ventures, and Enza Capital.

Founded in 2023, Lua is positioning itself at the forefront of a shift in enterprise AI, from tools that support workers to systems that perform work independently.

From assistance to execution

For years, enterprise AI has largely focused on augmentation. Tools embedded within platforms from Microsoft, Google Cloud, and Amazon Web Services have helped employees summarise documents, draft responses, and generate code.

Even automation platforms such as UiPath have tended to optimise existing workflows rather than replace them.

Lua is targeting a more fundamental shift.

Its platform allows organisations to create autonomous agents capable of handling multi-step workflows, such as customer onboarding, loan processing, and insurance claims without passing tasks between departments or systems.

“We’re in the race to shape how human-agent collaboration gets defined globally,” says co-founder Lorcan O’Cathain. “Organisations will be blends of humans and AI agents collaborating.”

Breaking the workflow model

Traditional enterprise systems rely on decomposition. A single process, such as a loan application is divided into verification, risk assessment, approval, and onboarding, each handled separately.

Lua argues that advances in AI remove the need for that fragmentation.

Its agents can take a request and carry it through multiple stages independently: collecting data, verifying documents, applying rules, and only escalating to humans when uncertainty or risk thresholds are exceeded.

Crucially, these systems operate through existing business channels, including WhatsApp, email, and voice, rather than requiring companies to adopt entirely new interfaces.

Early traction in financial services

Lua’s early deployments are concentrated in Kenya’s financial services sector, where manual processes continue to slow operations.

In many cases, unsecured retail loans can take between three and five days to process, largely due to manual know-your-customer (KYC) checks and document verification.

By automating these workflows, Lua aims to reduce turnaround times and eliminate bottlenecks.

The startup is also seeing early experimentation in how organisations structure teams around the technology. Some users are testing models where small teams supervise more than ten AI agents handling operational tasks.

A broader shift in enterprise AI

Lua’s approach reflects a wider transition in enterprise technology.

A study by PwC suggests that 79% of US companies are actively exploring AI integration, though most remain in early stages focused on experimentation and productivity gains.

Lua is targeting what comes next: moving from experimentation to execution.

“A year or two ago, AI could summarise documents and generate text, but it couldn’t reliably execute multi-step workflows with real business logic,” O’Cathain says. “That changed fast.”

The company frames its platform as turning large language models into “fully functioning employees” within an organisation, an indication of how AI is increasingly being mapped onto organisational structures rather than treated as standalone tools.

Built for local realities

Lua’s product design has been shaped by its operating environment.

In Kenya, enterprise software adoption is often driven by immediate operational challenges rather than long-term transformation strategies. As a result, the company has prioritised use cases with clear, measurable outcomes, faster onboarding, reduced processing times, and fewer manual queues.

The platform’s integration with widely used communication tools, including WhatsApp and other messaging services, reflects how businesses in the region already operate.

Lua also emphasises giving clients direct control over their AI agents, a feature it considers essential in markets where trust in third-party software providers is still developing.

Talking Points

It is notable that Lua is pushing beyond the current wave of AI tools by focusing on execution rather than assistance, building systems that can carry out entire business processes, not just support them.

This shift positions the startup at the frontier of enterprise AI, where the real value lies not in productivity gains alone, but in fundamentally redesigning how work is done across organisations.

At Techparley, we see how this model could be particularly impactful in African markets, where operational inefficiencies, especially in sectors like financial services create a strong case for end-to-end automation.

Lua’s decision to deploy agents through familiar channels such as WhatsApp and email is a practical advantage. It lowers adoption barriers and aligns with how businesses already communicate and operate on a daily basis.

The early focus on financial services is also strategic. Processes like loan approvals and customer onboarding are not only repetitive but time-sensitive, making them ideal candidates for automation with measurable outcomes.

Lua’s emphasis on integrating with existing systems rather than replacing them is a pragmatic approach. It allows companies to adopt the technology incrementally without disrupting core operations.

——————-

Bookmark Techparley.com for the most insightful technology news from the African continent.

Follow us on Twitter @Techparleynews, on Facebook at Techparley Africa, on LinkedIn at Techparley Africa, or on Instagram at Techparleynews.