SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Ozow is a fintech company redefining digital payments in South Africa and beyond. This role is part of a team building AI agents that automate real work across operations, support, fraud, risk, and engineering workflows.
You will build, test, and ship agent components alongside senior engineers, learning how to deploy AI safely in production within a regulated payments environment.
Key Responsibilities:
Agent Delivery: Ship at least one agent feature to production within your first 6 months with evaluation tests in place. Work with product, operations, and risk teams to understand the processes agents will support. Bring forward ideas for new automation use cases backed by the problems they solve.
Build and Integrate: Build and maintain agent workflows including tool calling, retrieval (RAG), prompt design, and multi-step task orchestration. Connect agents to internal APIs and data sources through secure, well-scoped integrations. Write clean, tested code that passes team review standards.
Quality, Reliability and Safety: Write evaluation tests that check agent outputs for accuracy, reliability, and safety before release. Monitor agents in production, triage failures, and feed findings back into improvements. Explain failure modes and how they're caught. Handle personal and financial data in line with POPIA and Ozow policy.
Tool Utilisation and Change Management: Document how each agent works, what it can and cannot do, and where human sign-off occurs. Follow Ozow's change, release, and security processes for every AI feature. Help business users adopt new agent tools through clear guides and demos.
Automation and AI: Use AI coding assistants responsibly to speed up delivery. Keep up with agent frameworks, MCP (Model Context Protocol), and model releases, sharing useful insights with the team. Flag manual processes that are good candidates for automation.
Requirements:
- Degree in Computer Science, Engineering, or related field, or equivalent practical experience
- Solid Python and comfort working with REST APIs and JSON
- Hands-on experience building with at least one LLM API, demonstrated through a project, GitHub repo, or production work
- Basic understanding of RAG, embeddings, and prompt engineering
- Git and working knowledge of Docker
- Careful approach to data; understanding why personal and financial data needs protecting
Nice to Have:
- Exposure to an agent framework (LangGraph, CrewAI, or agent SDK) or MCP
- AWS experience
- C# or .NET (much of the platform runs on ASP.NET Core)
- SQL (SQL Server or PostgreSQL)
- Interest in payments, fraud, or financial services
