Staff AI Engineer

Summary

Join Wizeline, a global AI-centric technology solutions provider, as a Staff AI Engineer. You will design and deliver AI-powered conversational workflows, handling millions of user interactions. Responsibilities include the full agent development lifecycle, integrating agents with various systems, and orchestrating LLMs. You will also implement safety and compliance measures and continuously optimize agent performance. Collaboration with cross-functional teams is crucial. The ideal candidate possesses extensive software engineering experience, LLM expertise, and a strong understanding of AI security.

Requirements

• 5 + years software engineering in production environments (TypeScript/Node, Go, or Python) with strong API design, testing, and debugging habits
• Hands-on LLM experience (GPT, Claude, Gemini, or fine-tuned OSS models): prompt engineering, function/tool calling, model selection, and evaluation
• Retrieval-Augmented Generation (RAG) : designing embedding pipelines and semantic search with Pinecone, Weaviate, etc
• Agent orchestration expertise : LangChain/LangGraph, OpenAI Agents SDK, CrewAI, or equivalent frameworks for tool-augmented reasoning loops
• ML-/DevOps literacy : containerisation, CI/CD, GPU/AI infra, caching, automated evals for nondeterministic systems
• Experience enforcing security guardrails & PII protection in user-facing AI products

Responsibilities

• Agent design & delivery: Translate business goals into deterministic, guard-railed conversational workflows; ship agents that handle millions of user interactions with high CSAT and low latency
• Full Agent Development Life-Cycle: Own scoping, design, implementation, automated evaluation/regression testing, deployment, rollout, monitoring, and iterative tuning
• Tool & data integration: Connect agents to enterprise APIs, vector/RAG knowledge bases, event streams, and internal micro-services so they can act , not just chat
• LLM orchestration & planning: Implement multi-model pipelines (e.g., comprehension → planning → critique → response) using frameworks such as OpenAI Function Calling, LangChain/LangGraph, or similar
• Trust, safety, and compliance: Implement multi-layer guardrails (rules + supervisor models), PII detection/masking, credential vaulting, and auditable traces that satisfy SOC 2 / GDPR-class requirements
• Continuous Optimisation : Inspect reasoning traces, analyse failure modes, improve prompts/skills/knowledge, and raise autonomous resolution rates week over week
• Cross-functional collaboration: Work closely with client PMs, CX leaders, and non-technical stakeholders; explain AI trade-offs clearly and gather feedback to shape roadmaps

Preferred Qualifications

• Real-time/voice agent delivery (e.g., SSE or WebSockets)
• Familiarity with emerging multi-agent protocols (MCP, A2A) or supervisor/critique agent patterns
• Domain experience in e-commerce, fintech, telecom, or large-scale customer support
• Prior work with automated prompt/agent regression testing frameworks

Benefits

• Competitive compensation & total rewards
• Health benefits & wellness programs
• Savings & retirement plans
• Global mobility opportunities
• Flexible work policy and remote-friendly approach
• Happy hours, gaming tournaments, sports activities & more
• Continuous learning & training programs with WizeAcademy
• Free certifications in cloud technologies and coding languages