prism-ai

byAman Khokhar

I want to build an AI-powered Mock Interview platform. Before writing code, analyze this idea and create the product requirements, user flows, system architecture, database design, AI agent architecture, and implementation plan. Do not start implementation until I approve the plan.

LandingInterview PracticeAuthenticationFeedback Review
Landing

Comments (0)

No comments yet. Be the first!

System Requirements

System Requirement Document
Page 1 of 7

prism-ai System Requirements Document

Introduction

This document outlines the system requirements for the "prism-ai" project, an AI-powered Mock Interview platform designed to help job seekers practice interviews and receive feedback. The document includes product requirements, user flows, system architecture, database design, AI agent architecture, and an implementation plan.

System Overview

Page 2 of 7

Product Interpretation and Delivery Boundary

  • Domain Descriptors:

    • AI-driven interview simulation: The platform simulates interviews using AI to provide realistic practice for job seekers.
    • Feedback and improvement tracking: The platform provides feedback to help job seekers improve their interview skills.
  • What is being built:

    • An AI-powered Mock Interview platform with a custom UI for job seekers to interact with the AI Interview Agent.
    • The platform includes distinct pages for Interview Practice, Feedback Review, Landing, and Authentication.
  • Excluded Assumptions:

    • Generic job application features
    • Traditional recruitment processes
  • Access:

    • Application-owned identity for job seekers to access their persistent interview and feedback records.
  • Behavior Contracts:

    • Conduct Mock Interview: Triggered by job seekers initiating a mock interview session, resulting in a simulated interview session and interview performance data.
    • Provide Feedback: Triggered by the completion of a mock interview session, resulting in a feedback report and improvement suggestions.
    • Initiate Mock Interview Session: Triggered by job seekers choosing to start a mock interview, resulting in an interview session initiation.
    • Review Feedback and Suggestions: Triggered by job seekers completing a mock interview session, allowing them to review feedback and suggestions.
Page 3 of 7

Page Content and Component Coverage

Landing Page

  • Purpose: Orient and acquire job seekers before they enter the protected mock interview and feedback workflows.
  • Components:
    • Introduction to the platform
    • Call-to-action for account creation or login
    • Overview of platform features and benefits

Authentication Page

  • Purpose: Provide account sign-up and login for job seekers.
  • Components:
    • Sign-up form
    • Login form
    • Password recovery option

Interview Practice Page

  • Purpose: Initiate and conduct mock interviews with the AI agent.
  • Components:
    • Interview session initiation
    • AI interview interface
    • Real-time feedback during the interview
Page 4 of 7

Feedback Review Page

  • Purpose: Review and receive feedback after a mock interview.
  • Components:
    • Feedback report
    • Improvement suggestions
    • Option to provide feedback on the session

Functional Requirements as Story Points

  • As a Job Seeker, I should be able to sign up for an account to access the platform. required_inference
  • As a Job Seeker, I should be able to log in to my account to access my interview sessions and feedback. required_inference
  • As a Job Seeker, I should be able to initiate a mock interview session to practice my interview skills. explicit
  • As a Job Seeker, I should be able to conduct a mock interview with the AI Interview Agent. explicit
  • As a Job Seeker, I should be able to review feedback and suggestions after completing a mock interview session. explicit
  • As an AI Interview Agent, I should be able to conduct mock interviews and provide feedback. explicit

User Personas

  • Job Seekers: Primary users who will use the platform to practice interviews.
  • AI Interview Agent: The AI agent that conducts mock interviews and provides feedback.

Core User Flows

Page 5 of 7

Job Seeker Flow

  1. Landing Page: Job Seeker visits the landing page to learn about the platform.
  2. Authentication Page: Job Seeker signs up or logs in to access their account.
  3. Interview Practice Page: Job Seeker initiates and conducts a mock interview.
  4. Feedback Review Page: Job Seeker reviews feedback and suggestions after the interview.

AI Interview Agent Flow

  1. Interview Practice Page: AI Interview Agent conducts the mock interview when initiated by the Job Seeker.
  2. Feedback Review Page: AI Interview Agent provides feedback and improvement suggestions after the interview.

Visuals Colors and Theme

  • Primary: #0056b3
  • Primary Light: #337ab7
  • Secondary: #f0ad4e
  • Accent: #d9534f
  • Highlight: #5bc0de
  • Background: #f7f7f7
  • Surface: #ffffff
  • Text: #333333
  • Text Muted: #777777
  • Border: #dddddd
Page 6 of 7

Signature Design Concept

The design concept for the prism-ai platform focuses on a clean and professional interface that emphasizes ease of use and accessibility. The landing page will feature a welcoming hero section with a brief introduction to the platform's capabilities, encouraging job seekers to sign up or log in. The interface will use a restrained color palette to maintain a professional appearance, with clear typography and intuitive navigation.

Interaction Model & Motion Direction

  • Landing Page: Animated with moderate scroll-triggered reveals and hover transitions to engage users.
  • Authentication Page: Static to prioritize clarity and ease of use.
  • Interview Practice Page: Animated with real-time feedback during the interview.
  • Feedback Review Page: Static with clear presentation of feedback and suggestions.

Non-Functional Requirements

  • Security: Secure access control for user accounts and data privacy. required_inference
  • Performance: The platform must handle concurrent interview sessions efficiently. required_inference
Page 7 of 7

Tech Stack

  • Frontend: React for Web
  • Backend: Python, FastAPI
  • Database: MySQL or MariaDB, use Alembic for migrations
  • AI Models: GPT 5.4 for user-friendly responses
  • AI Tools: Litellm for LLM Routing, Langchain
  • Local Orchestration: Docker, docker-compose
  • Server-side Orchestration: Kubernetes

Assumptions and Constraints

  • The platform is designed for job seekers to practice interviews and receive feedback.
  • The platform will not include generic job application features or traditional recruitment processes.

Glossary

  • AI Interview Agent: The AI component responsible for conducting interviews and providing feedback.
  • Job Seeker: A user of the platform seeking to improve their interview skills through practice sessions.
Landing design preview
Interview Practice: Job Seeker initiates mock interview session
Interview Practice: 1. AI Interview Agent conducts mock interview
Interview Practice: 2. Session could not be initiated
Feedback Review: 1. AI Interview Agent provides feedback report and improvement suggestions
Feedback Review: 2. Feedback could not be generated
Landing design preview
Interview Practice: Job Seeker initiates mock interview session
Interview Practice: 1. AI Interview Agent conducts mock interview
Interview Practice: 2. Session could not be initiated
Feedback Review: 1. AI Interview Agent provides feedback report and improvement suggestions
Feedback Review: 2. Feedback could not be generated