1 day ago
Bucharest, RomaniaStaff+
Responsibilities
- Architect and govern production-grade agentic systems featuring multi-agent orchestration, RAG pipelines, policy-based routing, memory management, and lifecycle observability.
- Define RAG standards covering chunking, embeddings, quality benchmarks, and documented metric-backed tradeoffs.
- Establish multi-LLM integration standards, including provider abstraction, fallback routing, and cost governance across OpenAI, Anthropic, Vertex AI, and open-source models.
- Own programme-level LLMOps, including evaluation strategy, prompt governance, observability standards, safety monitoring, and cost controls.
- Lead client architecture sessions, proof-of-concept delivery, and alignment between technology leadership and delivery teams.
- Publish reusable engineering patterns, accelerators, and standards for future client engagements.
- Define and communicate agentic-system quality metrics covering accuracy, latency, safety, cost, and business impact.
- Manage, develop, and performance-manage engineers, including setting individual development plans and conducting career conversations.
Requirements
- Production software engineering experience.
- Hands-on experience designing and deploying agentic AI solutions in production is required.
- Production-depth experience with LangGraph, CrewAI, AutoGen, or an equivalent agentic orchestration framework.
- Production experience calling OpenAI, Anthropic, or Vertex AI APIs, including provider abstraction, token management, and latency and cost tradeoffs.
- Ownership of RAG pipelines involving embeddings, chunking strategies, vector databases, and context engineering.
- Experience with LLMOps fundamentals such as evaluation harnesses, prompt versioning, and production observability.
- Cloud-native engineering experience with Kubernetes, Docker, microservices, serverless systems, CI/CD, and infrastructure as code using Terraform or Helm.
- Strong Python skills; Java or an equivalent backend language is acceptable.
- Production debugging and observability experience.
- Experience managing, developing, and performance-managing engineering teams.
- Candidates who have shipped three production agentic systems in four years are preferred over candidates with passive AI exposure.
Benefits
- Vendor fellowship access inside Anthropic, OpenAI, Microsoft, and Google engineering teams.
- Direct pathway to the Forward Deployed Engineer programme.
- Opportunity to work across every industry, enterprise technology stack, and level of organizational complexity.
Tech Stack
Categories
Forward Deployed
About Accenture
Accenture is a global professional services firm providing management consulting, systems integration and technology, cybersecurity, and business process outsourcing for enterprises and governments. It operates a services-driven model delivering projects and managed services, often with major cloud and software partners, across industries worldwide. Headquartered in Dublin and publicly traded on the NYSE (ACN), it originated as Andersen Consulting and adopted the Accenture name in 2001.