🏆 MHacks 2024 Best Developer Tool Award

GenHint: Guiding, Not Giving

VS Code AI Coding Assistant that Outputs Step-by-Step Guidance, Not Direct Solutions

Our Mission

We don't hand you the fish; we teach you how to fish.

Peihao Li*, Zichen Zhang*, Tongyuan Miao*, Yuchen Huang*

University of Michigan-Ann Arbor

{peihaoli, zhangzzc, tymiao, yuchenh}@umich.edu

* denotes equal contribution

This work was submitted as an MHacks24 project

Demo Video

Why GenHint?

Imagine a student struggling late into the night, toggling between countless browser tabs, endlessly searching for solutions that never seem to fit just right.

They’re overwhelmed, lost in the maze of documentation, forums, and scattered code snippets that offer more confusion than clarity.

Or perhaps they turn to ChatGPT/GitHub Copilot/Claude/Gemini, copying code that works but leaves them with no deeper understanding of what it does or why it’s written that way.

The result? A generation of learners bypassing the learning process, missing out on the essential skills that coding truly teaches—problem-solving, critical thinking, and the joy of figuring it out.

GenHint, our AI Coding Assistant, can help create a world where students don’t just get to the answer—they understand the path to it, becoming more skilled, confident, and independent coders. We believe that learning should be a guided, interactive experience, where every challenge is an opportunity to grow, every mistake is a chance to learn, and every line of code written is a step closer to mastery.

What we offer

GenHint transforms the way of learning coding in 3 aspects:

Structurize: A clear, step-by-step guidance that walks you through complex coding tasks, breaking them down into manageable actions.
Elaborate: If you need more understanding of any step, just click, and GenHint dives deeper, offering detailed explanations that clarify the “how” and “why” behind each move, making sure you truly grasp the concepts.
Review: Once your code is complete, GenHint reviews it, checking for accuracy and best practices, giving you feedback to ensure your code is indeed working.

How GenHint Works with Groq Fast Inference:

GenHint empowers you by providing step-by-step guidance right within your coding environment. Simply highlight a comment in your IDE and press a shortcut on your keyboard, and our GenHint AI Agent (based on Groq's fast inference architecture) will step in to help.

Based on in-context demonstrations, a system prompt that defines its role and user prompt (comments) , GenHint generates tailored "TODOs" comments, that are instantly inserted back into your workspace because of the fast inference powered by Groq.

GenHint supports 3 types of tasks: structurizing, elaborating, and reviewing. More details about the three pipelines are illustrated below.

Our goal is to enhance your learning by guiding you through the process (with steps, instructions, comments, etc) without handing over direct code. With GenHint, you're not just getting answers—you’re learning how to find them, fostering deeper understanding and skill development with every step.

Break a task into functions with serveral TODOs:

Elaborate on a TODO with detailed comments:

Review finished codes and provide indirect suggestions:

A Comprehensive Walk-through

VS Code Extension

Our project are fully deployed on VS Code Extension Marketplace.

Check it out here!! [Link]

Implementation Details

Please refer more information on our Github Page.

[Repository Link]

GitHub - zichenzhang04/GenHint: GenHint: an AI coding assistant that helps newbie developers with pseudocode hints and progress trackingGenHint: an AI coding assistant that helps newbie developers with pseudocode hints and progress tracking - zichenzhang04/GenHint

DevPost Website

Project Team

Peihao Li

He is a junior undergraduate at the University of Michigan, pursuing a double major in Honors Computer Science and Mathematics, with a minor in Music. His academic passion mainly lies in generating physics-based intelligence and reconstructing efficient and reliable world models. He is fortunate to work with Prof. Min Xu at Carnegie Mellon University and Prof. Hong Liu at Osaka University. In his spare time, he enjoys exploring National Parks, playing tennis and guitar.

Zichen Zhang

He is a junior undergraduate student at the University of Michigan, majoring in Computer Science. He is passionate about deep learning models, web technologies and entrepreneurship. He is interested in both the academia and the entrepreneurial world. He co-founded Collage, an EdTech startup to assist students with course scheduling. He previously served as a research assistant at the U-M Minji Lab and U-M Direct Brain Interface Lab (UMDBI).

Tongyuan Miao

He is currently a sophomore at the University of Michigan, pursuing a double major in Computer Science and Computer Engineering. A current member of the Michigan Investment Group, he is passionate in both the research of deep learning models, and their applications in the industry. He serves as a research assistant at the U-M Green D.E.I lab focusing on the implementation of deep learning models in Insect Behavioral Biology.

Yuchen Huang

He is an aspiring computational linguist and mathematician, currently a sophomore at the University of Michigan, pursuing a double major in Computer Science and Mathematics. He serves as a research assistant at the Computational Neurolinguistic Lab with interests in neuro-LLM alignment, while also working as a teaching assistant for Honors Multivariable and Vector Calculus.

Acknowledgement

We would like to extend our sincere thanks to our sponsors for their generous support.

Page updated

Report abuse