EY GDS Hackpions 2.0

2387 Registered Allowed team size: 1 - 3

This campaign is over.

idea phase
starts on:
Mar 11, 2021, 12:30 PM
ends on:
Apr 11, 2021, 06:29 PM
starts on:
Apr 16, 2021, 02:30 PM
ends on:
May 27, 2021, 06:25 PM



------Shortlisted teams qualified for Semi-Finale------

Theme Team Name
Intelligent visual identity check
Chatbot for interview scheduling
AR/VR for virtual events
NLP-based tagging solution
Intel Information Risk Analyzer
NCU Hackers
Employee emergency app (IoT)

Fostering creativity and growth through innovation

Welcome to EY GDS Hackpions 2.0, the third edition of the hackathon from the Enablement Services function of EY Global Delivery Services (GDS) — the hottest virtual event of 2021!

The hackathon aims to connect participants with experts, to tackle real-world challenges. It brings together technology champions, students and working professionals such as yourself, from varied domains. Through the course of the event, we’d encourage you to gather unique ideas and innovation, that can help alleviate business problems faced by EY GDS teams during and after the pandemic. Join this unique and fun upskilling experience, to showcase your coding craftsmanship.

In this virtual hackathon event, we’ll be asking you to develop tools, apps, bots and other advanced solutions. You can partner with hackers from across the world and build hacks in critical and extensive domains, by leveraging technologies such as artificial intelligence, cloud and data analytics. You’ll also get a chance to be mentored by professionals from EY GDS and HackerEarth, who will handhold you while you take up the challenge to turn hacks into real-world business solutions.

What does the challenge include?

You are expected to ideate on problem statements and churn out advanced solutions. Your prototype should belong to any one of the following technology stacks: App development, API development. data science, artificial intelligence or machine learning, game development, WebApp development, full stack and cloud computing.

Why join the event?

You and your team can compete virtually from wherever you are, irrespective of whether you are a student, a new joiner or an experienced working professional. This is an opportunity to work with and learn from the industry’s leading minds, and to develop solutions, using the tools and resources of EY GDS.

Top teams, winners and all the finalists stand a chance to win exciting prizes and rewards, along with an opportunity to work for EY GDS as interns, based on the panel’s recommendation.

NOTE: EY GDS participants (mentors) must use their official email IDs to sign-up.


Guidelines for building your prototype solution

The expected statistical solutions for all six themes, are mentioned below: 

  • The solution should ideally be suitable for deployment in an Azure cloud platform.
  • The tools can leverage open source, custom libraries, application programming interfaces (API), from cloud platforms. 
  • Azure and MS Tech stack continue to be the most preferred stacks, over which we would like the solutions to be built. Open source technologies can be used if Azure and MS Tech stack cannot solve the problem, or if you don’t have access to the above-mentioned stacks. However, you must design the solution in a way that it can be easily ported to Azure and MS Tech stack and provide a valid business reason for choosing open source technologies, during the evaluation stage.
  • The preferred technology stack is C#, .NET and Python for artificial intelligence (AI) or machine learning (ML). You are also free to use libraries that are not licensed under Affero General Public License (AGPL).
Intelligent visual identity check


The Visual Identity team manually scrutinizes the quality of each document, to ensure that they are aligned with the brand guidelines. This is a tedious and time-consuming task.

You need to build an intelligent text-recognizer solution, to automate this process and verify documents, against the given guidelines.


Propose and design a solution that will carry out the following tasks:

  • The intelligent visual identity checker should allow the user upload a pdf or a jpeg file as the input.
  • In the input file, the solution should identify text, font face, text color, font size, text placement (pixel or X–Y coordinates), background color, bullet symbol, bullet size, bullet color and bullet indentation level.
  • The solution should cross-check the file against given brand guidelines, and generate an output file, marking any exceptions.


A solution (e.g., a web-app, API or tool) that can accurately identify the text details and typography, and cross-check it against the provided guidelines.


A sample folder with the input file, output file and guidelines, for verification purposes.

Chatbot for interview scheduling


Scheduling interviews on behalf of Recruiting/Hiring managers involves multiple people such as interviewers/hiring managers, recruiters and candidates. This means finding common availability is a challenge and a time-consuming activity. This includes getting calendar availability of hiring managers, booking rooms in case of in-person interviews, and ensuring invites have the right links for virtual interviews. 

You need to build a conversational AI chatbot solution, that will help with scheduling interviews faster, saving time and effort. 


Propose and design a solution that will carry out the following tasks:

  • Develop a conversational AI bot solution, that interacts with hiring managers to communicate, converse and confirm their availability. This will facilitate a hiring manager’s participation in an interview, for a given date, time or location.
  • The bot should be able to provide a list of common free slots, based on the calendar availability of all the participating managers, in an interview. (Click here to see the sample of a hiring manager’s calendar.)
  • Based on the interview scenarios,the AI bot should be able to understand the responses of the hiring managers, and respond to the confirmations received on availability, rejection and requests to re-check for other time and date slots. It should also pull in additional EY recruitment team members, in case the hiring manager asks specific questions, that are beyond just scheduling.


An AI assistant that makes appointments.


sample folder with the calendar of a hiring manager, some general chat scenarios and responses on bot assistance provided to the candidate, and an ideal bot structure.

NLP-based tagging solution


The Intelligent Automation team manually processes a large volume of exception data. The manual process includes examining the exception reasons and categorizing them broadly into business or technical issues, which is time consuming and increases turnaround time. 

You need to build an intelligent solution, to auto label exceptions as they occur.


Propose and design a machine learning algorithm, that will carry out the following tasks:

  • Use training data and keywords across categories, for each of the exceptions, as given below:
  • Create an input excel file, with a list of exceptions.
    • Business exception
    • System exception
    • Other
  • Generate an output text file, with the respective exception tags.
  • Create a self-learning model to recognize patterns, determine correlation with other input data set features, learn from data and tag exceptions as they come.


An AI assistant that identifies exceptions in real time.


sample folder with the input data, training data and keywords for tagging the exception type, and output with business or system tags.

AR/VR for virtual events


As the world shifts toward remote working, events are now conducted virtually. We seek help in creating a platform or app that would enable users to attend virtual events, providing an interactive environment for participants, with the help of augmented reality (AR), virtual reality (VR) and gamification. 

You need to build an AR/VR mobile app, that ensures a truly immersive, seamless experience across various devices for individuals attending virtual events.


Propose and design a solution that will carry out the following tasks:

  • Create an AR/VR app for hosting or attending virtual events.
  • Create virtual events like conference or meeting hall, tech booths, quizzes, polls, surveys, games, learning sessions and award ceremonies.
  • The solution may include gamification like audio, video, chat box, motion sensor or calibration (mobile app), showing a map of the event layout or activities, etc.
  • The app can include interactive features like face recognition for event registration, breakout room or booths, quizzes to score redeemable points or likewise.
  • Add additional options like speech- or voice-to-text conversion for data capturing, voice-operated navigation and login, as a host or an attendee.


An app that creates and manages virtual events seamlessly.


sample event for creating AR/VR scenarios.

Intel Information Risk Analyzer


Information received continuously over emails from various intel sources (mentioned below in the “Deliverables” section) provide details about the incidents that can have a potential impact on the organization. 

We require a text analyzer solution that can read and classify these emails, to understand the potential impact of the incident on the organization, enabling quicker response through proactive monitoring. 

Using technologies like artificial intelligence and machine learning, the analyzer should produce dynamic risk scores, after factoring the organizational data with the incident information. Then, the analyzer should send an event escalation or early warning, based on historical information and patterns. 


Propose and design a text analyzer solution (app, dashboard or tool) that will carry out the following tasks:

  • The text analyzer takes emails (in a pdf format) as the input.
  • From the input mail content, the text analyzer categorizes each email as one of the four source types, based on the email sender’s domain address, as listed below: 
    • Core: @controlrisks.com
    • ISOS: @internationalsos.com
    • VCC: @everbridge.net
    • Risk advisory: @riskadvisory.net
  • The text analyzer categorizes each email under one of the ten risk types, based on the keywords.
  • The country name gets extracted from the email body.
  • A table tracking the total count of inputs, risks, sources, incident severities and countries, gets generated.
  • Information from the intel sources are mapped with EY information, to produce country level dynamic risk score by category. 
  • The text analyzer evaluates the historical information for a country based on the event type and provides an impact score basis the escalation matrix for the system, to trigger an early warning signal.
  • Historical trends of the impact on geography are provided, that indicate seasonality of incidents.


A web-app interface, to upload multiple pdf files or emails and produce output tables with a count of — input files, risk type, source type and table of the identified countries with their count — for each header.


A sample folder consisting of the input pdf files, a list of risk categories, a list of source categories with respective keywords and the list of countries that need to be identified.

Futuristic perspective

The dashboard may show the country’s risk profile, based on different risk categories. It may show an early warning signal, historical trend and impact analyses, and recommend probable actions based on the event type.

Employee emergency app (IoT)


The Emergency Response Team (ERT) is responsible for ensuring safe evacuation of employees, during an emergency. 

You need to build an employee emergency app or dashboard, to enable real-time evacuation and enhance safety levels, in the organization.


Propose and design an app, that supports the corresponding features, for the three interface logins, as given below:

  • Command center: This interface has the following features:
    • Send alert and push notifications to ERT and employees as “evacuation alert”.
    • Provide a map view, to check exit points, routes, etc. with a zoom option.
    • Check live location and count of employees marked “safe”.
    • Enable an ERT chat box, with suggested messages like “all clear,” “man missing,” “emergency,” etc.
  • ERT: This interface has the following features:
    • Provide an option to mark attendance (as “present” or “on-leave”).
    • Share live location with command center.
  • Employee: This interface has the following features:
    • Share live location with command center.
    • Get “evacuation alert” from command center and check safety instructions.
    • Direct toward nearest exit point, using map-view.
    • Track exit and entry points, via appropriate geo-fencing technology, using Bluetooth, GPS, mobile tower triangulation, BLE Beacon, etc.*
    • Enable an SOS button for extreme emergency alerts to be sent to ERT, when trapped or incapacitated.
    • Enable an option to mark people “safe”.


An app that enables an easy emergency response planning, control and evacuation. 


sample folder with a floor map and location-wise details.

*Note: The cost of any hardware components used to build and provide a demo of the solution will be reimbursed, if your idea gets shortlisted.

Prizes INR 10,50,000/- in prizes

Main Prizes
Finale top winners (3)
INR 1,50,000/-

Three teams out of six finalists

Finale runners up (3)
INR 1,00,000/-

Three teams out of six finalists

Second runners up (6)
INR 25,000/-

One team per theme and six teams out of 12 eliminated semi-finalist teams (18 semi-finalist teams in total)


Social Share

View All Notifications