Agentic AI Engineer Intern

Docusign • San Francisco, CA

2d • Hybrid

About The Position

Docusign is seeking a curious Agentic AI Engineer Intern to help build autonomous AI workflows and multi-agent systems. The intern will work with Generative AI technologies and frameworks such as CrewAI, Python, SQL, and Gemini ADK, collaborating with senior engineers to design, deploy, and evaluate autonomous agents for complex reasoning, tool use, and task execution. The three-month internship provides hands-on experience with LLMs, prompt engineering, agent orchestration, and evaluation frameworks. This is an individual contributor role reporting to the Sr. Manager, Data & Analytics.

Requirements

• Currently pursuing a Bachelor's or Master's degree in Computer Science, Data Science, AI/ML, or a related technical field.
• Graduating in December 2027 or May/June 2028.
• Experience with Python and fundamental data structures.
• Experience with SQL and relational database concepts.
• Experience with Large Language Models (LLMs) and core concepts such as tokenization, context windows, and function calling/tool use.
• Experience with CrewAI or similar agentic frameworks, such as LangChain or Google ADK.
• Must already be authorized to work in the United States on a full-time, permanent basis without current or future sponsorship; the role is not eligible for CPT, OPT, or sponsorship.

Nice To Haves

• Hands-on experience with Google ADK or Google Cloud Vertex AI platforms.
• Prior experience writing automated tests or using LLM evaluation tools such as Ragas, DeepEval, Phoenix/Arize, or custom evaluation harnesses.
• Familiarity with vector databases, embeddings, or RAG architectures.
• Active open-source contributions or personal side projects involving AI agents.

Responsibilities

• Design, build, and test collaborative AI agent workflows using CrewAI, Gemini ADK, and Python.
• Integrate Google Gemini ADK and other modern LLM APIs for reasoning, function calling, and structured outputs.
• Connect agents to SQL databases, APIs, and internal tools for real-world execution and retrieval-augmented generation (RAG).
• Craft, benchmark, and refine system prompts, agent roles, and task descriptions to improve task success rates and reduce hallucinations.
• Assist in establishing quantitative evaluations and measurement frameworks to benchmark agent performance, accuracy, latency, and cost efficiency.
• Monitor agent execution traces, debug multi-turn agent conversations, and implement fallbacks for robust behavior.

What This Job Offers

• Job Type: Full-time
• Career Level: Intern