Techfest 2024-25

AI Quest

Background:
AIQuest is an internal knowledge-sharing platform designed to facilitate question-and-answer
interactions within organizations, similar to Stack Overflow. It enables employees to ask
questions, provide answers, and incorporates AI-generated replies to improve efficiency. AI
Quest combines user-generated content with official responses to build a comprehensive
knowledge base that fosters collaboration and knowledge sharing.

The AIQuest platform can serve as a valuable resource for knowledge sharing and collaboration
within the organization. Below is a high-level overview of the tasks you need to perform as part of
this solution.

Tasks:

1. Platform Setup

● Develop a question-and-answer platform where employees can ask questions and

provide answers.
● The platform should include:
○ A clean and intuitive user interface.
○ Functionalities to categorise and tag questions.
○ User profiles to manage contributions and engagement.
○ Integration with internal tools like wikis and GitHub repositories for reference.

2. User-Generated Content (Questions and Answers)

● Implement functionality for employees to post questions related to work.

● Allow other employees to answer these questions, forming a collaborative knowledge
base.
● Track and store questions and answers in a database for future reference and AI learning.

3. Official Answers

● Designate roles for subject matter experts or department representatives who can provide
official answers.
● Allow official answers to be marked distinctly from user-generated answers for clarity.
● Ensure that these authoritative answers appear prominently to maintain accuracy.
4. AI-Generated Replies

● Integrate AI models capable of generating potential replies based on existing content and
internal knowledge resources.
● Train the AI to provide contextual and relevant responses by analyzing previous
questions, answers, and official content.
● Ensure that AI responses are clearly differentiated from human-generated replies.

5. Voting and Ranking System

● Implement a voting mechanism where employees can upvote or downvote answers.

● Build an algorithm that ranks answers based on their usefulness and accuracy.
● Ensure that the most upvoted or verified answers appear at the top to provide easy
access to high-quality information.

6. Moderation and Quality Control

● Introduce a moderation system where content can be reviewed for quality, accuracy, and
appropriateness.
● Allow moderators to flag or remove inappropriate content and verify official answers.
● Implement automated checks for spam or irrelevant content.

7. Continuous Improvement and Analytics

● Implement an analytics dashboard that tracks platform usage, including the number of
questions asked, answers provided, and AI-generated responses.
● Analyze user feedback and engagement patterns to improve the platform.
● Refine the AI model and add new features based on usage insights and user feedback.

8. Notification and Alert System

● Notify users when their questions are answered or when relevant topics are updated.

Workflow:

1. User asks a question → The question is posted on the platform.

2. Other users can provide answers → Employees contribute their expertise.
3. AI generates potential replies → AI suggests answers based on existing knowledge.
4. Official answer is marked → Designated experts or department heads provide verified
responses.
5. Users upvote and rank answers → The most helpful answers rise to the top.
6. AI learns from interactions → The system continuously improves from user interactions
and feedback.
 7. Search functionality → Employees search for questions and answers using keywords.
8. Notifications and alerts → Users receive updates on question activity or new content.
9. Analytics provide insights → Administrators gain insights into platform usage and
effectiveness.

Implementation Considerations:

1. User-Friendly Interface: Prioritize simplicity and usability in the design.

2. Robust Search Capability: Ensure that users can easily find relevant content with
advanced filtering options.
3. Integration with Existing Systems: Seamlessly integrate with internal knowledge
repositories like GitHub, internal wikis, etc.
4. Data Privacy and Security: Implement strong authentication and encryption measures to
protect sensitive organizational information.
5. Third-Party Software and Data: Participants are not allowed to use third-party software
or data applications. All solutions must be built using the provided resources or tools
developed in-house. Violation of this rule could lead to disqualification.
6. Custom Sample Data: Participants are free to use their own sample data for AI training,
but it must remain relevant and aligned with the scope of the problem statement.
Submissions with data significantly deviating from the problem scope may be rejected.

STRUCTURE:
There will be three stages to the competition -

Round 1: Abstract review meet

For abstract submission, registered teams must create a presentation with no more than 15
slides. The presentation should include a detailed plan of action, the underlying logic,
implementation details, and the software and online platforms that will be utilized in the project..
Shortlisted teams will showcase their work and improvements will be suggested by the experts

Round 2: Final review meet

The updated report will be showcased based on which 5 teams will be shortlisted for the final
presentation.

Round 3: Presentation and Working Demonstration

Working demonstration and presentation of the solution
Top 5 teams shortlisted from Stage 2 will present their solution at Techfest 2024-25

ABSTRACT SUBMISSION:

● Prepare a report and send it to [email protected]

● Subject of the mail: AIQuest_TeamID_TeamLeaderName
● Mention all your team details clearly in the mail.
TEAM SPECIFICATIONS AND ELIGIBILITY

● All students with a valid identity card from their respective educational institutions are eligible to
participate in the competition
● A team can have a maximum of 4 members.
● Students from different institutes can be a part of the same team.

GENERAL RULES

1. Every team has to register online on our website for the competition. A Team ID will be
allocated to the team on registration which shall be used for future references.
2. A team can register at any point of time before 19th October 2024 and submit the final
abstract.
3. The decision of the organizers or judges shall be treated as final and binding on all.
Techfest has all the rights to verify the identity and accuracy of the details provided by the
participants.
4. No responsibility will be held by Techfest, IIT Bombay for any late, lost or misdirected
entries.
5. The idea presented by the teams should be original (not protected by means of
patent/copyright/technical publication by anyone else).
6. Note that at any point of time the latest information will be that which is on the website.
However, registered participants will be informed through mail about any changes on the
website.
7. All modes of official communication will be through the Techfest e-mail.

Timeline

Last date of Registration 27th October Participants need to register

before this date

Last date for abstract 27th October Submission of First draft

submissions report

Round 1 4th November Experts will review your

abstract and guide you for
improvements

Round 2 11th November Participants will showcase

updated report and 5 teams
will be shortlisted

Round 3 17th-19th December Final Live presentation in

front of Judges
CERTIFICATE POLICY

1. Top three teams in the grand finale will be awarded Certificate of Excellence.
2. E-Certificate of participation will be given to the teams scoring more than the critical
marks which will be decided later.

PRIZE MONEY
The Prize money will be awarded to the top 3 Winners via NEFT and will be processed within
30 working days after receiving the Prize Money from Sponsors.

The Winners have to mail the following information (immediately after the announcement of
the results) to [email protected] .

Subject: Competition name, team id - your position (example- AIQuest, AIQuest-2309234 – 1st
position)
Body of mail-
1. Account Holder’s Name
2. Account Number
3. Bank name and Branch name
4. IFSC Code
5. Photograph of Bank Passbook as a proof