20 hours ago
Bucharest, RomaniaMid Level
Responsibilities
- Implement production AI agents, prompts, tools, retrieval pipelines, and orchestration logic.
- Integrate AI components with enterprise APIs, applications, databases, queues, events, and workflow services.
- Build automated tests, evaluation datasets, and instrumentation for agent behavior and reliability.
- Diagnose model, retrieval, tool-use, and integration failures.
- Implement agent capabilities including tool calling, structured outputs, state, memory, hand-offs, guardrails, retries, human-in-the-loop steps, and deterministic workflow nodes.
- Implement RAG pipelines using embeddings, vector or hybrid search, metadata filters, reranking, and evaluation datasets.
- Contribute to secure coding, documentation, peer review, releases, agile ceremonies, demonstrations, and backlog refinement.
- Support proposals and client workshops, contribute reusable engineering assets, coach colleagues, and participate in continuous learning.
Requirements
- At least 3 years of experience in software, data, or AI engineering.
- Strong Python or comparable programming skills, API development experience, and version-control experience.
- Practical experience with LLM applications, RAG, agents, embeddings, and structured outputs.
- Hands-on experience building agents with a production-oriented framework such as LangChain, LangGraph, Microsoft Agent Framework, Semantic Kernel, OpenAI Agents SDK, AutoGen, CrewAI, or an equivalent.
- Practical experience with FastAPI or similar API frameworks, Pydantic or comparable schema validation, asynchronous programming, Git, and automated testing.
- Experience implementing RAG pipelines, tool calling, agent state and memory, hand-offs, guardrails, retries, human-in-the-loop steps, and deterministic workflow nodes.
- Familiarity with MCP, enterprise API integration, queues or events, Docker, Kubernetes, or managed application platforms.
- Ability to instrument agent executions using tools such as LangSmith, MLflow, Langfuse, OpenTelemetry, or platform-native equivalents.
- Ability to collaborate iteratively with product, architecture, data, and user-experience specialists.
Tech Stack
Categories
About PwC
PwC is a global network of professional services firms providing assurance, tax, and advisory work for businesses and public-sector clients. It sells project-based and managed consulting, audit, and deals services, and implements and integrates business applications. Formed in 1998 by the merger of Price Waterhouse and Coopers & Lybrand, it is headquartered in London and is one of the Big Four accounting firms.