Most fintech AI is a feature. We are building an intelligence layer.

 

Payble is the financial and business operating system for the world's underserved majority, the 400 million micro-enterprises across Africa and the global south that traditional finance was never built for. We have already built POS payments, business banking, inventory management, invoicing, and analytics. The foundation is in place.

 

What we build next defines everything. We are looking for one exceptional person to architect and own Payble's AI layer, the intelligence that makes every merchant feel like they have a CFO, a compliance officer, a business analyst, and a financial advisor in their pocket. One that understands their business, speaks their language, and gets smarter every single day.

 

THE PROBLEM WE ARE SOLVING

A provision store owner in Lagos has no CFO. A fabric trader in Kano has no compliance officer. A smallholder farmer in Ebonyi has no business analyst. They are running businesses with the same complexity, cash flow pressures, and regulatory obligations as formal companies, entirely without infrastructure, data, or institutional support.

Payble's AI layer will change that. Not by giving them a dashboard. By giving them an intelligent system that watches their business, understands their patterns, anticipates their risks, and acts or prompts them to act before problems become crises.

The same intelligence, applied with local context, serves an SME owner in London, a market trader in Accra, a logistics operator in Nairobi. This is not an African product. This is a product for everyone that incumbent finance forgot. That market is in the billions.

WHAT YOU WILL OWN

You are not joining a team. You are founding the AI function. That means you will design the architecture, make the foundational decisions, and ship the first intelligent experiences, all of which will shape how Payble works for millions of users.

Concretely, you own four things:

• The merchant intelligence layer: A unified, real-time context model representing everything Payble knows about each merchant: sales patterns, cash flow cycles, inventory behaviour, compliance position, and financial health. This is the foundation everything else sits on.
• The agentic workflow system: The agents that monitor merchant state, identify moments requiring action, and either act or surface the right prompt at the right moment. Reorder alerts. Invoice reminders. Compliance deadlines. Cash flow warnings. Business insights that a market trader would otherwise never have access to.
• The AI experience surface: How intelligence appears inside Payble. The language, the timing, the interaction model. You are responsible not just for whether recommendations are correct, but whether they land in a way that a first-generation business owner can act on at 7am when they are opening their shop.
• The feedback and improvement loop: The infrastructure that tells you whether the AI is working, so that Payble's intelligence compounds continuously rather than decaying after launch.

THE TECHNICAL MANDATE — FIRST 6 MONTHS

We are not asking you to research. We are asking you to ship. These are the systems we need built, in order of priority:

• Merchant context model: A structured, continuously updated representation of each merchant's business state, built on Payble's existing transaction, inventory, invoicing, and banking data. The data architecture that makes everything else possible.
• Agentic workflows on high-confidence signals: Inventory reorder triggers, invoice payment chasers, cash flow gap warnings, and compliance deadline monitors. Event-driven, precise, and designed to be trusted by users whose livelihoods depend on them.
• Compliance intelligence engine: Mapping merchant financial activity against FIRS, CBN, and UK HMRC rule sets, surfacing obligations before they become penalties, and walking merchants through resolution steps they would otherwise need an accountant to explain.
• Business intelligence layer: Turning transaction history and sector benchmarks into merchant-specific insight. Margin analysis. Peak trading patterns. Comparative performance. The kind of insight that only large retail chains with analytics teams currently have access to.

WHAT WE NEED FROM YOU

We will be direct. This profile is rare. We know that. We are looking globally because we believe the right person exist and we are willing to wait for them rather than compromise.

Non-negotiable capabilities:

• Production LLM engineering at genuine depth: You have shipped AI features using foundation model APIs in a live product used by real people. Not demos. Not prototypes. You have solved hallucination, latency, and failure mode problems in an environment where being wrong has consequences.
• Agentic systems architecture: You have designed and built agents that monitor state, execute multi-step reasoning, handle failures gracefully, and know when to escalate to a human. You understand orchestration frameworks and have opinions on their tradeoffs grounded in real experience.

Strongly valued:

• Financial services domain knowledge: You understand how small businesses manage cash, why compliance feels threatening, and how informal businesses think about tax obligations. You design AI that feels like it was built by someone who understands their world.
• Mobile-first engineering experience: Payble's primary surface is mobile. The AI experience needs to be designed for a 5-inch screen, often in low-connectivity environments, used by someone who may have limited digital literacy.
• Multilingual or low-literacy UX thinking: You have thought about how intelligent systems communicate across language and literacy barriers, and you have opinions about how to make AI feel accessible rather than intimidating.
• Experience in African or other high-growth emerging markets: You understand the infrastructural, behavioural, and trust dynamics that shape how technology is adopted in these environments.

WHAT YOU WILL NOT FIND ANYWHERE ELSE

You will not find a mandate this broad at any company that is not building something genuinely new.

 

At Payble, the AI Product Engineer does not implement a roadmap written by someone else. They write the roadmap. They make the foundational architecture decisions. They own the intelligence layer of a product that will serve millions of people who have never had access to the kind of financial intelligence that incumbents take for granted.

 

The equity on offer reflects that. This is a founding-level role in a company with a genuine global ambition, one that has already demonstrated traction, secured regulatory certifications, and built the product foundation on which the AI layer will sit.

THE PERSON

Beyond the technical specification, we are looking for someone with a specific orientation toward the work:

HOW TO APPLY

We are not running a standard application process. We are looking for a specific person, and we will know them when we find them.

Send the following to joy@payble.io with the subject line: AI Product Engineer - Your Name

• A brief account of the most technically complex AI system you have shipped to production, what it did, what broke, what you fixed, and what you would do differently.
• Your view on the single most important architectural decision in building a merchant intelligence layer for a fintech operating in low-data, high-trust-risk environments.
• Links to work you are proud of, could be on GitHub, writing, projects. We read them.

There is no CV screening stage. If your response to the above is compelling, we will schedule a conversation with the founder within 5 business days.