Hybrid Senior Software Engineer bei WebMD

About the Role

Roles and Responsibilities

• Architect and build AI-powered tools using LLMs (e.g., OpenAI, Claude, Mistral, etc.).
• Integrate and customize AI agents using frameworks like LangChain, CrewAI, or custom solutions.
• Collaborate with product and research teams to translate requirements into prompt workflows and agent behaviors.
• Design and implement scalable systems and infrastructure to support AI agents and workflows, using suitable tools and technologies.

• Design advanced, modular prompt chains that incorporate memory, dynamic context injection, and multi-step reasoning across agent workflows.
• Experiment with and optimize agentic architectures for real-world tasks (retrieval, generation, summarization, etc.).

• Collaborate with UI/UX or front-end teams to integrate AI-powered features into user-facing applications as needed.

• Willingness and ability to quickly learn and adapt to new technologies, frameworks, and tools as needed. Open to working across the stack to support the development of AI-powered solutions. Comfortable stepping beyond core skill sets to contribute wherever needed in the AI application lifecycle.

• Write unit and integration tests to ensure reliability.
• Participate in code reviews, CI/CD pipelines, and deploy models/services using best practices.

• Upskill peers in GenAI, prompt engineering, and best practices.
• Contribute to knowledge-sharing and architectural reviews across teams.
• Stay updated with evolving AI ecosystems, tools, and research.

Requirements

• Strong programming fundamentals (OOP, security, performance).
• Experience building scalable and modular systems using modern frameworks and tools.
• Proficiency with LLMs, OpenAI/Anthropic APIs, LangChain, Hugging Face Transformers, or similar.

• 3+ years of experience working with GenAI and LLMs.
• 2+ years of experience in prompt engineering and AI agent architectures.
• Familiarity with agent frameworks like CrewAI, Autogen, Cursor, or Windsurf.
• Strong understanding of the Cursor Rule (intent → context → action), a principle for structuring agent behavior, and real-world experience applying it in workflows.
• Solid understanding of how to construct and pass dynamic context to LLMs, including techniques like session memory and prompt templating.
• Experience adapting or customizing LLMs is a plus.

• Familiarity with front-end frameworks or relational databases is a plus.

• Experience with Git, CI/CD workflows, and test automation.
• Familiarity with DevOps tools, monitoring, and logging in production environments.

• Excellent communication and collaboration skills.
• Self-driven and eager to mentor and share knowledge.