AI & Automation Intern (Prompt Engineering + Workflow Automation)

ChaiOne•Houston, TX

2d•Hybrid

About The Position

The AI & Automation Intern will join the AI Innovation team and help design, build, and optimize AI-powered systems involving language, voice, vision, and automation. The role includes writing prompts for large language models, creating automated agent workflows, designing conversational voice experiences, and integrating multimodal AI into business processes. The intern will work on production-style AI agents that combine LLMs, APIs, automation platforms, and conversational design to address real-world problems. The internship is described as a summer or semester opportunity with a flexible duration and potential to extend. It is remote or hybrid if local, and compensation is unpaid or provided as academic credit depending on the program.

Requirements

Currently enrolled in, or strongly interested in, a program related to Computer Science, AI, Engineering, Cognitive Science, Linguistics, or a similar field.
Strong written and verbal communication skills.
Genuine interest in prompt engineering, AI workflows, and conversational systems.
Familiarity with large language models such as OpenAI, Claude, or Gemini.
Comfort working with APIs, webhooks, and tools such as Postman.
Familiarity with automation tools such as Zapier, n8n, AutoGen, or LangGraph.

Nice To Haves

Exposure to Python or JavaScript.
Experience with frameworks such as LangChain.
Coursework or interest in computer vision or multimodal AI.

Responsibilities

Write, optimize, and refine prompts for large language models such as GPT-4, Claude, and Gemini.
Build multi-step prompt chains for complex tasks and workflows.
Tune prompts for accuracy, safety, tone, and business context.
Build end-to-end automation pipelines using no-code or low-code tools.
Connect APIs, data sources, and services to support AI-driven workflows.
Integrate LLM-powered agents into business processes, including error handling and edge-case design.
Design natural, goal-oriented voice conversations, including greetings, intent clarification, confirmations, fallbacks, and closings.
Write natural, conversational, latency-optimized voice prompts and handle unclear inputs gracefully.
Consider when LLMs should not be used, including high-risk or compliance-sensitive situations.
Build conversational flows using node-based visual flow editors.
Connect voice agents to backend systems such as CRMs, databases, and APIs to trigger actions.
Monitor call analytics and optimize flows based on real user behavior.
Build simple vision-model workflows for tasks such as object detection or quality checks.
Combine vision, language, and automation into multimodal AI agents.
Test and quality-assure AI systems across text, voice, and multimodal outputs.
Continuously refine workflows to improve accuracy, safety, and outcomes.