AI Engineer Intern

AI Engineering Intern — AR5 Labs

Remote | Internship

ABOUT AR5 LABS

AR5 Labs builds AI-powered automation tools for data analytics. Our flagship product, PlotStudio AI, is a desktop analytics platform where specialized AI agents collaborate through a multi-stage pipeline to deliver the kind of analysis a senior data scientist would produce, complete with code, visualizations, statistical rigor, and executive-grade narrative. Think Cursor, but for data analytics.

Product: https://www.plotstudio.ai/

LinkedIn: https://www.linkedin.com/company/ar5-labs/

ABOUT THE ROLE

We're looking for an AI Engineering Intern to help build and improve the AI agent pipeline behind PlotStudio AI.

This is not a research internship. You'll work on production AI systems: prompt engineering, agent orchestration, model evaluation, and the infrastructure that makes multi-agent workflows reliable. You'll see how AI agents are actually built, tested, and shipped to real users.

You'll work directly with the founder and the AI/ML team. The codebase is small enough that your contributions will ship to production quickly, and complex enough that you'll learn how a real multi-agent system works from the inside.

If you're a student who wants to build AI systems, not just study them, this is the role.

RESPONSIBILITIES

- Help design, test, and iterate on prompts for PlotStudio's multi-agent pipeline (routing, planning, coding, interpretation agents)

- Build evaluation scripts to benchmark agent output quality across different datasets and question types

- Run experiments comparing model performance (GPT-4.1, GPT-5.1, Claude, Gemini) on specific analytical tasks

- Help improve agent self-correction logic: how the system detects errors, retries, and adjusts its approach

- Work on domain skill modules that inject specialized knowledge (finance, healthcare, econometrics) into the agent pipeline

- Build internal tooling for tracking agent accuracy, token usage, and failure modes

- Help extend the agent orchestration layer (built with LangGraph) to support new workflow types

- Document prompt versions, experiment results, and engineering decisions as you go

QUALIFICATIONS

Required:

- Currently pursuing a degree in Computer Science, Data Science, AI/ML, or a related field (rising sophomore, junior, or senior)

- Comfortable writing Python. You should be able to read, write, and debug code independently

- Basic understanding of how LLMs work: prompts, tokens, context windows, temperature

- Genuine interest in AI agents, not just chatbots. You should have an opinion on why single-prompt AI falls short for complex tasks

- Willingness to experiment, break things, and iterate quickly

- Self-directed and comfortable asking questions when you're stuck

- Able to commit 15-20 hours per week

Nice to have:

- Experience using the OpenAI API, Anthropic API, or any LLM SDK

- Familiarity with LangChain, LangGraph, CrewAI, or any agent framework

- Personal projects involving LLMs, prompt engineering, or AI agents

- Coursework in machine learning, NLP, or statistics

- Familiarity with Git and version control

- Experience with FastAPI, Flask, or any Python backend framework

- Understanding of statistical methods (regression, hypothesis testing, time series)

WHAT YOU'LL LEARN

- How a production multi-agent system is built: routing, planning, execution, self-correction, and interpretation agents working in sequence

- Prompt engineering at a level most people never see: versioned prompts, domain-aware injection, structured output parsing, and quality gates between agents

- How to evaluate AI systems beyond vibes: building benchmarks, measuring accuracy, tracking regressions across prompt versions

- How to ship AI to real users in a small, fast-moving team

WHY THIS INTERNSHIP

- You'll work on a real multi-agent AI system, not a tutorial project

- Small team means high visibility and direct mentorship from the founder

- Your work will ship to production and be used by real customers

- You'll have something concrete for your resume: "I helped build and evaluate the AI agent pipeline for a production analytics platform"

MUST HAVE

We are across the United States and Canada. Must be a citizen of either country.

HOW TO APPLY

Follow our LinkedIn page and DM us directly, or email us:

1. A short intro about yourself