AI Engineer

Skip to main contentSign InAI Engineer page is loadedAI EngineerApplylocationsOsaka, Japantime typeFull timeposted onPosted 15 Days Agojob requisition idJR1343About UsAutomation Anywhere is the leader in Agentic Process Automation (APA), transforming how work gets done with AI-powered automation. Its APA system, built on the industry’s first Process Reasoning Engine (PRE) and specialized AI agents, combines process discovery, RPA, end-to-end orchestration, document processing, and analytics—all delivered with enterprise-grade security and governance. Guided by its vision to fuel the future of work, Automation Anywhere helps organizations worldwide boost productivity, accelerate growth, and unleash human potential.QUALIFICATIONSEducation: Bachelor’s or Master’s in CS, AI/ML, Data Science or equivalent practical experienceExperience: 4–7 years in software engineering; 3+ years building production-grade automation solutionsDemonstrated end-to-end delivery of agentic AI systems or complex enterprise RPA prototypesCertifications (Preferred): AWS, Azure or GCP; RPA platforms such as UiPath or Automation AnywhereScope: Leads solution architecture independently, owns LLM adaptation strategy end-to-end, mentors junior engineers; leads stakeholder discovery workshopsSKILLSAgentic AI: LangChain/LangGraph, AutoGen, CrewAIAgent patterns: tool use, memory, multi-agent coordination, guardrails, failure recoveryLLM Fine-Tuning & Adaptation: LoRA/QLoRA with HuggingFace PEFT or Unsloth; Dataset prep, evaluation benchmarking, model versioning; Serving fine-tuned models:vLLM,GPTQ, GGUFRAG & Vector Infrastructure: Pinecone, Weaviate, Qdrant; embeddings, retrieval evaluationRPA: UiPath, Automation Anywhere, Power Automate in productionEngineering: Python (production quality); Cloud AI services (Bedrock,Azure, OpenAI, Vertex AI)RESPONSIBILITIESAgent Design & Engineering: Architect multi-agent systems with branching logic, exception handling & human-in-the-loop escalation.Define agent tool integrations, memory, context management & state persistence.LLM Adaptation Strategy:Own fine-tuning strategy (fine-tune vs RAG vs prompt engineering) and deliver end-to-endManage GPU training runs, model merging, quantization & production servingRPA & HYBRID AUTOMATION:Build RPA task bots as execution layers within agentic workflowsArchitect AI agent ↔ RPA handoff logic and exception managementProduction & Operations:Build agent evaluation frameworks; implement observability & tracing (LangSmith, Arize)CI/CD for agent/model deployments; diagnose hallucination, tool misuse & cost runawayStakeholder & Leadership:Lead use-case discovery workshops; communicate architecture trade-offs to non-technical audiences All unsolicited resumes submitted to any @automationanywhere.com email address, whether submitted by an individual or by an agency, will not be eligible for an agency fee.Loading