In Nairobi, the pitch for AI is rarely about the model. It's about the dashboard that a manager can actually read, the forecast that works with local sales data, and the workflow that doesn't require a PhD to adjust. Keja Analytics, a small consultancy of two to ten people, is building its practice on that exact premise [LinkedIn, 2024]. The firm, founded by AI/ML engineer Ken Mbaya, offers end-to-end data and AI solutions, from no-code implementations to secure analytics dashboards, with a focus on custom delivery for business clients [Keja Analytics, 2024]. There are no press releases about funding rounds or unicorn valuations here. The traction signal is simpler: a sustained offer to translate complex technology into operational advantage for companies that may be buying it for the first time.
The services-led wedge into enterprise AI
Keja's positioning is deliberately not that of a SaaS company. Its website describes a consulting and solutions provider that helps businesses "use technology to reduce costs, prevent lost revenue and gain a competitive advantage" [Keja Analytics, 2024]. The wedge is services, not software. For a market where data maturity can vary wildly even within a single industry, a custom-built dashboard or a tailored demand forecasting model can be a more pragmatic entry point than a standardized product. Founder Ken Mbaya, described as a highly skilled AI/ML engineer with over three years of experience building such systems, leads the technical delivery [F6S, 2024]. The team includes at least one other engineer, Jakinda Oluoch, who holds the title of AI and Data Engineer [F6S, 2024]. Their public work centers on automating data workflows and designing business intelligence interfaces, the unglamorous plumbing that makes AI insights actionable.
Why a boutique model makes sense now
The bet rests on a few intersecting trends. First, the proliferation of cloud infrastructure and accessible machine learning tools has lowered the technical floor for building custom solutions. Second, as global AI hype reaches East Africa, local businesses are seeking guidance they can trust from providers who understand regional commercial rhythms and data constraints. A small, agile team can move faster and build closer to specific client needs than a large multinational consultancy or a distant SaaS vendor. The model is inherently low-overhead and project-based, which aligns with the undisclosed or bootstrapped funding structure suggested by the lack of any public investment rounds [Keja Analytics, 2024]. Growth is tied to reputation and project referrals, not a burn rate.
The team and its capacity
The company's capacity is defined by its founder-led technical core. Ken Mbaya's background as a Data & AI Consultant and AI/ML Engineer forms the foundation of Keja's offerings [LinkedIn, 2024]. His profile indicates a focus on practical implementation, a necessary skill for a services business. The supporting technical team, while small, suggests the ability to handle multiple concurrent client engagements or complex, multi-phase projects.
| Role | Name | Background Note |
|---|---|---|
| Founder, AI/ML Engineer | Ken Mbaya | Data & AI Consultant; 3+ years building custom analytics and ML systems [F6S, 2024]. |
| AI and Data Engineer | Jakinda Oluoch | Technical team member focused on data engineering and AI implementation [F6S, 2024]. |
Where the model meets its limits
A pure services business faces inherent scaling challenges. The primary constraint is people; revenue is a direct function of billable hours and project throughput. Without a product to scale, growth is linear. The competitive set is also broad and fragmented.
- Global SaaS incumbents. Tools like Microsoft Power BI, Tableau, and various forecasting SaaS offer low-cost, self-serve options that could appeal to more tech-savvy clients.
- Larger consultancies. Established regional and global IT services firms have deeper benches and broader relationships, which can be decisive in larger enterprise deals.
- In-house teams. As talent becomes more available, some companies may opt to build internal data capabilities, cutting out the external consultant.
Keja's differentiation must be its deep customization, client intimacy, and speed. The risk is that this remains a successful but niche practice, unable to capture the kind of scalable value that attracts institutional investment.
The ideal customer profile here is a Nairobi-based small or medium enterprise in sectors like retail, logistics, or manufacturing. This is a company that has outgrown spreadsheets, recognizes the value in its data, but lacks the internal skills or bandwidth to build a solution. They need a trusted partner to deliver a working system, not just a software license. For Keja, competing means staying closer to that client's specific operational reality than any off-the-shelf tool ever could. The realistic competition isn't another startup; it's the decision to do nothing, or to hire a full-time data scientist. Keja's bet is that its tailored, project-based approach is the most cost-effective path between those two poles.
Sources
- [Keja Analytics, 2024] Keja Analytics homepage | https://www.kejaanalytics.com/
- [LinkedIn, 2024] Keja Analytics LinkedIn Company Page | https://www.linkedin.com/company/keja-analytics/
- [F6S, 2024] F6S profile of Ken Mbaya | https://www.f6s.com/member/kenmbaya
- [F6S, 2024] F6S profile of Jakinda Oluoch | https://www.f6s.com/member/jakindaoluoch