SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Payoneer, a global financial platform founded in 2005, is seeking an AI System Architect to design, build, and deploy production-grade AI agents that automate and augment customer journey workflows across support, operations, onboarding, disputes, and KYC processes.
In this role, you will:
• Design and ship LLM-powered AI agents that automate customer-facing and internal workflows, with a focus on production quality and reliability.
• Partner with cross-functional stakeholders (Customer Journey, Product, Tech, Risk, Compliance) to scope problems, prioritize use cases, and align agent solutions with business objectives.
• Develop agents end-to-end: prompt engineering, tool/function calling, retrieval-augmented generation (RAG) over internal knowledge bases, memory management, multi-step reasoning, and graceful fallback mechanisms.
• Integrate agents with internal systems including CRM/ticketing platforms, knowledge bases, data platforms, and payment systems via REST APIs and event-driven pipelines.
• Establish evaluation frameworks and quality gates: offline evaluations, A/B testing, human-in-the-loop review, tracing, and observability to ensure agents are accurate, safe, and policy-compliant.
• Champion responsible AI practices: build guardrails for PII protection, fraud detection, compliance, and hallucination mitigation; ensure traceability, auditability, and safe escalation to humans.
• Drive adoption across Customer Journey teams: enable users, gather feedback, iterate rapidly, and measure business impact through CSAT, average handle time, deflection rates, accuracy, and cost-to-serve metrics.
• Stay current with the agentic AI ecosystem (models, orchestration frameworks, evaluation tools) and bring proven practices into Payoneer's architecture.
Requirements:
• Bachelor's degree in Computer Science, Software Engineering, Data Science, Industrial Engineering, or related quantitative/engineering discipline.
• 3+ years of hands-on experience building, deploying, and maintaining production-grade software using Python, TypeScript/Node.js, or similar languages.
• 2+ years of practical experience developing and deploying LLM-powered applications and AI agents, including prompt engineering, tool/function calling, RAG, and agent orchestration.
• Hands-on experience with at least one LLM or agent development framework (LangChain/LangGraph, LlamaIndex, OpenAI Agents SDK, Anthropic SDK, Semantic Kernel, AutoGen, or equivalent).
• Strong experience designing and implementing REST APIs, webhooks, and event-driven integrations connecting multiple services and data sources.
• Experience with SQL, embeddings, vector databases, and building retrieval pipelines across structured and unstructured data.
• Experience deploying applications to cloud environments (AWS, GCP, Azure) and working with CI/CD pipelines. Familiarity with LLM evaluation, monitoring, observability, and safety mechanisms is a plus.
• Hands-on, execution-oriented approach with a track record of taking solutions from concept to production in fast-paced, iterative, 0-to-1 environments.
• Strong problem-solving and analytical thinking skills with an experimentation-driven approach to debugging and improving AI solutions.
• Ability to translate business needs into technical solutions, make sound engineering trade-offs, and connect technical decisions to measurable business outcomes.
• Strong communication skills to explain technical concepts, agent behavior, limitations, and trade-offs to non-technical stakeholders.
• Proactive, detail-oriented, and comfortable managing multiple technical workstreams while collaborating cross-functionally.
• Experience working in international environments and fluency in English.
• Curiosity and enthusiasm for emerging AI technologies and best practices in production AI development.
• Familiarity with customer support platforms or contact center tools is an advantage.
