Want to know if this job is worth applying to?
大阪市, 日本
Onsite
About Us
Automation 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.
QUALIFICATIONS
Education: Bachelor’s or Master’s in CS, AI/ML, Data Science or equivalent practical experience
Experience: 4–7 years in software engineering; 3+ years building production-grade automation solutions
Demonstrated end-to-end delivery of agentic AI systems or complex enterprise RPA prototypes
Certifications (Preferred): AWS, Azure or GCP; RPA platforms such as UiPath or Automation Anywhere
Scope: Leads solution architecture independently, owns LLM adaptation strategy end-to-end, mentors junior engineers; leads stakeholder discovery workshops
SKILLS
Agentic AI: LangChain/LangGraph, AutoGen, CrewAI
Agent patterns: tool use, memory, multi-agent coordination, guardrails, failure recovery
LLM Fine-Tuning & Adaptation: LoRA/QLoRA with HuggingFace PEFT or Unsloth; Dataset prep, evaluation benchmarking, model versioning; Serving fine-tuned models:vLLM,GPTQ, GGUF
RAG & Vector Infrastructure: Pinecone, Weaviate, Qdrant; embeddings, retrieval evaluation
RPA: UiPath, Automation Anywhere, Power Automate in production
Engineering: Python (production quality); Cloud AI services (Bedrock,Azure, OpenAI, Vertex AI)
RESPONSIBILITIES
Agent 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-end
Manage GPU training runs, model merging, quantization & production serving
RPA & HYBRID AUTOMATION:
Build RPA task bots as execution layers within agentic workflows
Architect AI agent ↔ RPA handoff logic and exception management
Production & Operations:
Build agent evaluation frameworks; implement observability & tracing (LangSmith, Arize)
CI/CD for agent/model deployments; diagnose hallucination, tool misuse & cost runaway
Stakeholder & 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.
