Selecting the right teams is crucial for the success of a hackathon. However, the current process of shortlisting teams is often manual, time-consuming, and prone to human error. Moreover, the lack of a standardized evaluation process can lead to inconsistent results and a lack of transparency.
To develop innovative solutions that streamline the process of shortlisting teams for hackathons, we have come up with following features:
-
Evaluation Criteria :
-
Parsing technique is used which involves extracting and preprocessing relevant information from a resume using parser and NLP techniques.
-
Domain prediction which uses word embeddings to determine the most suitable domain for a user's skills and experience.
- Skills : 35% weightage
- Project : 25% weightage
- Work Experience : 20% weightage
- Competitions/Achievements : 20% weightage
-
-
Team Selection : Based on above techniques,candidate would be assigned resume scores which involves calculating similarity scores between the user's skills and job requirements using word embeddings, and using a weighted metric to evaluate the user's experience in different key areas. Teams with higher scores would have high preference of being shortlisted.
-
Tie : In case of tie between teams with same resume score,admin can conduct a test by sending a test link via mail to respective teams.Deserving teams can now be shortlisted easily based on test scores.
-
Further verification(if required) : In case if admin wants to further verify the teams,they can do so by having one-one high quality video conferencing with respective teams.Admin can also keep record of candidates in database and can view top 10 teams.
-
Candidate Analysis : Admin can have further real-time analysis of team by looking at bar graph generated based on resume scores and pie chart of thier domain(Eg: Blockchain).
-
Hackathon reminders : Admin can notify the registered candidates by posting updates on timeline.All necessary announcements can be done here.
-
User Features :
-
Candidate can check his resume score based on above techniques.
-
If they are looking for teammates, they can have hassle free communication using Open Forum(Chat Application) and grow their networking.
-
-
Recommendation features : For candidate to improve his/her resume score further,we have provided :
-
Recommended skills which he can learn further in future.
-
Recommended courses which can be taken up in order to learn the above recommended skills.
-
- Instead of using word-based embedding,we can use sentence-embedding to get proper context.
- Further parse profile via LinkedIn.
- Conducting end-to-end hackathon management system once the hackathon ends (eg: uploading the code,demo video files)
- Providing a payment gateway to send cash prizes to hackathon winners.