Inspiration
Difficult workplace conversations are something almost every professional faces, but very few people get a safe space to practice them. Whether it is asking for a salary raise, handling conflict with a manager, preparing for an interview, raising an HR concern, or resigning professionally, people often feel anxious and unprepared. We built NextMove AI to help users rehearse these moments before they happen, so they can communicate with more confidence, clarity, and emotional control.
What it does
NextMove AI is an AI-powered workplace conversation rehearsal studio. Users can choose a workplace scenario, describe their situation, and practice a realistic conversation with an AI manager, HR representative, recruiter, coworker, or client. The app supports text and voice-based rehearsal using speech-to-text and text-to-speech. Users can also select pressure levels such as Calm, Realistic, and Challenging to make the simulation more lifelike. After the session, NextMove AI generates a communication scorecard with professionalism, clarity, confidence, empathy, emotional control, and risk level, along with improved response suggestions and a before-vs-after comparison.
How we built it
We built NextMove AI using MeDo by first creating a detailed product requirements document that clearly defined the core action: helping a user practice a difficult workplace conversation through AI roleplay. MeDo helped generate the full application structure, including the landing page, dashboard, simulation room, conversation history, authentication flow, and feedback report experience. We then iterated through MeDo’s chat-based development workflow to improve the UI, add judge demo mode, enhance the AI simulation room, and integrate plugins such as Large Language Model, Text-to-Speech, Speech-to-Text, Text-to-Video, and Image-to-Video for a more immersive experience.
Challenges we ran into
The biggest challenge was keeping the product focused while still making it feel impressive enough for a hackathon. Workplace communication is a broad problem, so we had to avoid turning the app into a generic AI assistant. Instead, we focused on one strong workflow: rehearse a difficult conversation, receive feedback, and improve the response. Another challenge was making the AI roleplay feel realistic and scenario-specific instead of giving generic chatbot replies. We solved this by adding role-based behavior, pressure levels, judge demo mode, and structured feedback reports.
Accomplishments that we're proud of
We are proud that NextMove AI feels like a polished, real-world AI product rather than just a simple chatbot. The AI Simulation Room became the centerpiece of the experience, with immersive roleplay, voice interaction, pressure levels, and actionable feedback. We are also proud of the communication scorecard, which turns a conversation into measurable improvement areas. Most importantly, we built and deployed a full-stack, public AI application quickly using MeDo, showing how natural language development can turn an idea into a functional product.
What we learned
We learned that the quality of an AI-generated app depends heavily on the clarity of the brief. Instead of starting with a vague prompt, we defined the users, core action, data, pages, UI style, and V1 constraints before generating the app. We also learned that the best AI products are not just about generating answers, but about creating interactive experiences that help users improve. Through MeDo, we saw how fast iteration, plugin integration, and conversational development can dramatically speed up the process of building and refining a product.
What's next for NextMove AI
Next, we want to make NextMove AI even more personalized and career-focused. Future improvements include deeper emotion analysis, personalized communication coaching plans, role-specific training modules, interview preparation packs, manager training simulations, multilingual support, and progress tracking over time. We also plan to add organization-level features where companies can use NextMove AI to train employees in communication, leadership, conflict resolution, and customer handling.
Built With
- large-language-model-api
- medo
- medo-plugins
- openai-api
- speech-to-text
Log in or sign up for Devpost to join the conversation.