Staff AI Agent Engineer

Zendesk; Berlin; Hybrid; EUR 90.000 - 120.000; Vollzeit

Zusammenfassung

A leading software company in Berlin seeks a highly experienced Staff AI Agent Engineer to drive innovation in AI systems. You will design robust AI agents, mentor developers, and manage deployments in a hybrid work environment. Ideal candidates have a strong background in LLMs and experience with Python and cloud deployment.

Leistungen

Hybrid work model

Flexible schedule

Diversity and inclusion commitment

Qualifikationen

• Expert in LLM-Oriented System Design with strong prompt engineering.
• Mastery of Tool Integration and APIs in complex environments.
• Experience in designing, building, and optimizing robust RAG pipelines.

Aufgaben

• Architect and lead the development of robust AI agents using Python.
• Evaluate and select appropriate foundation models and services.
• Mentor and guide junior and mid-level developers.

Kenntnisse

LLM-Oriented System Design

Tool Integration & APIs

Retrieval-Augmented Generation

Performance Optimization

Programming & Tooling

Ausbildung

Ph.D. or Master's degree in Computer Science, AI, ML, or related field

Tools

Python

FastAPI

Cloud deployment (AWS/GCP/Azure)

Jobbeschreibung

Overview

The Agentic Tribe is revolutionizing the chatbot and voice assistance landscape with Gen3, a cutting-edge AI Agent system that is goal-oriented, dynamic, and truly conversational. Gen3 reasons, plans, and adapts to user needs in real-time using a multi-agent architecture and advanced language models to deliver personalized experiences and handle complex, "off-script" inquiries.

About the Role: We are seeking a highly experienced Staff AI Agent Engineer to drive innovation and technical leadership. You will design, develop, and deploy intelligent autonomous agents that leverage Large Language Models (LLMs) to streamline operations, shape the cognitive architecture for AI-powered applications, and guide other engineers. You will own cross-cutting technical initiatives, serve as a go-to expert for complex problems, and engage with stakeholders to influence strategy and execution.

Responsibilities

• Architect, design, and lead the development of robust, stateful, and scalable AI agents using Python and modern agentic frameworks (e.g., LangChain, LlamaIndex), setting technical direction and best practices for engineering teams.
• Strategize and oversee the integration of AI agent solutions with existing enterprise systems, databases, and third-party APIs to create seamless end-to-end workflows across the product, identifying and mitigating architectural risks.
• Evaluate and select appropriate foundation models and services from third-party providers (e.g., OpenAI, Anthropic, Google), analyzing strengths, weaknesses, and cost-effectiveness for specific use cases.
• Own and drive the entire lifecycle of AI Agent deployment, from concept to production and beyond for large, ambiguous, or highly complex initiatives—collaborating with product leadership and ML scientists to deliver effective agent solutions.
• Troubleshoot, debug, and optimize complex AI systems, ensuring performance, reliability, and scalability in production, and mentor other engineers in advanced problem-solving techniques.
• Define and continuously improve platforms and methodologies for evaluating AI agent performance, setting metrics, driving iterative improvements, and influencing industry best practices.
• Establish and enforce best practices for documentation of development processes, architectural decisions, code, and research findings to ensure knowledge sharing and maintainability across the team.
• Mentor and guide junior and mid-level developers, fostering technical excellence and contributing to career development.

Core Technical Competencies

• Expert in LLM-Oriented System Design: Architecting multi-step, tool-using agents (e.g., LangChain, Autogen) with strong prompt engineering, context management, and handling LLM quirks. Implement advanced reasoning patterns like Chain-of-Thought and multi-agent communication.
• Mastery of Tool Integration & APIs: Designing secure, scalable integrations of agents with external tools, databases, and APIs in complex environments.
• Retrieval-Augmented Generation (RAG): Designing, building, and optimizing robust RAG pipelines with vector databases, chunking, and hybrid search techniques.
• Leadership in Evaluation & Observability: Defining and implementing LLM evaluation frameworks and monitoring for latency, accuracy, and tool usage across production systems.
• Safety & Reliability: Designing defenses against prompt injection and robust guardrails with effective fallback strategies.
• Performance Optimization: Managing LLM token budgets and latency through routing, caching (e.g., Redis), and other optimization techniques.
• Planning & Reasoning: Designing agents with long-term memory and advanced planning capabilities (e.g., ReAct, Tree-of-Thought).
• Programming & Tooling: Proficiency in Python, FastAPI, LLM SDKs; experience with cloud deployment (AWS/GCP/Azure) and CI/CD for complex AI applications.

Bonus Points (Preferred Qualifications)

• Ph.D. or Master's degree in Computer Science, AI, ML, or related field.
• Comprehensive understanding of foundational ML concepts (attention, embeddings, transfer learning).
• Experience translating academic research into production-ready code.
• Familiarity with fine-tuning techniques (e.g., PEFT, LoRA).

The Interview Process

1. Initial Call with Talent Team – 15 mins
2. Interview with a Hiring Team member – 45 mins
3. Take-home technical challenge
4. Technical interview with two engineers for deeper discussion – 1 hour
5. Final interview with CTO or Engineering Leader – 45 minutes

About Zendesk

Zendesk builds software for better customer relationships. It empowers organizations to improve customer engagement and understand their customers. Zendesk products are easy to use and implement, enabling rapid innovation and scalable growth. More than 100,000 paid customer accounts in over 150 countries and territories use Zendesk products. Based in San Francisco, Zendesk operates globally.

Zendesk is an equal opportunity employer and is committed to diversity, equity, and inclusion in the workplace. We recruit without regard to race, color, religion, national origin, age, sex, gender, gender identity, gender expression, sexual orientation, marital status, medical condition, disability, military or veteran status, or any other characteristic protected by applicable law. By submitting your application, you consent to Zendesk collecting personal data for recruiting and related purposes. Zendesk’s Candidate Privacy Notice explains what data is processed, where, why, and rights related to this processing.

Hybrid: This role supports a hybrid work model with a mix of onsite and remote work. The specific in-office schedule is determined by the hiring manager. Zendesk offices around the world provide space for collaboration and learning, with flexibility for remote work part of the week.