AI Engineer

Responsibilities:

• You'll build the application layer of AI systems for enterprise clients. This means taking a technical architecture designed by the Architect and a data foundation built by the Data Integration Specialist, and building the interfaces, workflows, and automation logic that users actually interact with. Dashboards, chat interfaces, report generators, monitoring tools, Slack and WhatsApp integrations, automation workflows you build it all.
• You'll work in 2-week build sprints with client demos at the end of each sprint. You ship fast, get feedback, and iterate. The goal is working software in the client's hands within weeks, not a perfect system in months.

Requirements:

• 2–5 years experience in full-stack software engineering, with a tilt toward AI/ML applications or backend development
• Strong Python skills, this is the primary language for the AI layer
• Experience with LLM APIs (OpenAI, Anthropic, or open-source model serving)
• Familiarity with orchestration frameworks (LangChain, LlamaIndex, or custom RAG implementations)
• Frontend capability able to build clean, functional interfaces using React, Next.js, or similar
• Experience with low-code platforms (Make.com, n8n, Retool) for rapid prototyping and client-facing tools
• Comfort with cloud infrastructure (AWS, GCP, or Azure) for deployment and hosting
• Ability to ship fast without sacrificing reliability, pragmatic engineering over architectural perfectionis
• Experience building chat interfaces or conversational AI applications
• Familiarity with workflow automation for business processes (not just technical workflows)
• Previous work building tools for non-technical end users, people who need intuitive interfaces, not developer tools
• Experience with mobile-first design (many end users in our client contexts access tools primarily on phones)
• Experience building for African markets where connectivity, device constraints, and infrastructure reliability are real considerations

Engagement structure: Contract, project-by-project. Typical engagement: 8–12 weeks, with heaviest workload in weeks 4–10 (after the data foundation is in place and the architecture is set). Approximately 25–30 hours per week during active build phases. Remote-friendly. Must be available during GMT+1 to GMT+3 working hours.