PlacePilot – AI-Powered Placement Intelligence Platform

login page
portal
mock interview
profile page
aptitude test
coding practice
leetcode
skill gap
job roles
job management
analytics
admin dashboard
students details
resume analyzer

Inspiration

Many engineering students prepare for placements using separate platforms for coding practice, aptitude tests, resume building, and interview preparation. However, these platforms often provide separate scores without giving students a complete picture of their placement readiness. We were inspired to build a platform that helps students understand not only how they perform, but also why they need improvement, which skills they should prioritize, and what they should do next.

What it does

PlacePilot is an AI-powered placement preparation platform that evaluates student performance and provides personalized guidance based on a selected job role. Its key features include: Role-Specific Readiness Prediction: Uses student performance data to estimate readiness for a target job role. Realistic AI Mock Interviews: Simulates technical and HR interviews with AI-generated questions, answer evaluation, and relevant follow-up questions. Explainable AI: Uses SHAP to explain which performance factors influence the readiness prediction. Skill-Gap Analysis: Compares the student's current skills with the requirements of the selected role. Resume Analysis: Identifies relevant skills and areas for improvement. Coding and Aptitude Assessments: Tracks performance across different assessment areas. Personalized AI Recommendations: Generates learning suggestions and a structured four-week improvement plan. Continuous Progress Tracking: Updates the student's skill profile and readiness results as new assessment and interview data becomes available. Placement Officer Dashboard: Helps monitor student readiness and identify common skill gaps across a batch.

How we built it

We designed PlacePilot as a full-stack application using React for the frontend, FastAPI for the backend, and SQLAlchemy for database operations. SQLite supports local development, while PostgreSQL is planned for production deployment. The machine-learning component uses a Random Forest classifier to predict readiness categories: Ready, Moderately Ready, and Needs Improvement. SHAP is used to explain the model's predictions. Google Gemini API supports AI-powered interview evaluation and personalized recommendations, with a demo fallback when the API is unavailable. JWT authentication and password hashing support student and administrator access. The platform combines performance data from assessments, coding activity, resume analysis, and mock interviews to create a unified student skill profile. This profile is compared with the selected job role's requirements to identify skill gaps and generate personalized improvement plans.

Challenges we ran into

One of our main challenges was designing a unified approach to combine different performance measures, such as coding, aptitude, technical knowledge, resume quality, and interview performance. Another challenge was making the readiness prediction understandable rather than presenting it as an unexplained score. We addressed this by incorporating SHAP-based explanations and role-specific skill-gap analysis. We also considered the need for personalized recommendations and reliable updates whenever a student completes a new assessment or interview. For development and demonstration, synthetic student data and a fallback evaluation mechanism help us test the workflow before real-world validation.

Accomplishments that we're proud of

We are proud to have built PlacePilot, an all-in-one platform that combines machine learning, explainable AI, and generative AI to help students prepare for placements. Our key achievement is integrating readiness prediction, skill-gap analysis, AI mock interviews, and personalized improvement plans into one platform. We are especially proud that PlacePilot turns student performance data into clear, actionable guidance for continuous improvement.

What we learned

Through PlacePilot, we explored full-stack application development, REST API integration, database design, machine-learning workflows, model explainability, and generative AI integration. We also learned that a useful AI system should do more than produce predictions. It should explain its results, provide actionable guidance, and adapt when new information becomes available.

What's next for PlacePilot – AI-Powered Placement Intelligence Platform

Our long-term goal is to help students prepare more effectively and enable placement officers to make better-informed training decisions. PlacePilot aims to transform placement preparation from separate practice activities into a personalized, explainable, and continuously improving process.