We have hit our limit on the number of signups. Signups are closed as of 11/7/2025.

Challenge Overview: Enhanced Emissions Estimates for ESG Data Completeness

Many companies do not disclose comprehensive greenhouse gas emissions across Scopes 1, 2, and 3, resulting in material gaps within ESG datasets used for investment decision-making and regulatory compliance. This challenge focuses on developing a more accurate and reliable methodology to estimate emissions for non-reporting companies, with the objective of improving current backtesting performance (now in the high-70% range).

Participants will leverage a combination of internal and external data sources—such as revenue, headcount, industry classification, country, and ESG scores—to generate robust, scalable emissions estimates. All required datasets will be supplied for the challenge. To address privacy and MNPI considerations, the datasets will be obfuscated while preserving analytical utility.

Event Schedule:

Date Day Event Format Start Time End Time
21 Nov 2025 Friday In-Person (mandatory) 09:00 17:00
22 Nov 2025 Saturday Students may collaborate offline. Mentors might not be available.
23 Nov 2025 Sunday Students may collaborate offline. Mentors might not be available.
24 Nov 2025 Monday Virtual 09:00 11:00

About Fitch Group

Fitch Group: A Global Leader in Financial Information Services, Fitch Group operates in over 30 countries and is a prominent provider of financial information services. The group comprises three primary divisions:

  • Fitch Ratings: A global leader in credit ratings and research.
  • Fitch Solutions: A leading provider of credit market data, analytical tools, and risk services.
  • Fitch Learning: A preeminent training and professional development firm.

With dual headquarters in London and New York, Fitch Group is owned by Hearst.

Requirements

Participation and Team Structure

  • Participants will be organized into teams of four to five members.
  • Each team is expected to design and develop a conceptual solution that addresses the challenge.

Submission Requirements

  1. A URL to a public GitHub repo that will contain the following:
    • Data Science Notebook & README.md with detailed thought process on
      • Problem understanding and setting the Hypothesis
      • Exhaustive EDA-trying to understand the pain points in the data given
      • Data Engineering ,handling the messy data
      • Model Selection and your intuition behind the models selected and Experimentation
      • Model Tweaking and hyper parameter tuning shows the command over the model selected
      • Evaluation and how it ties down to the business
    • A submission.csv file with your predictions in the format matching the example in notebooks/submission.csv
    • Please make sure no passwords, api keys or access tokens etc. are checked into the GitHub repo.
    • Wireframes mockups or storyboards for the application if anything like that was produced (optional).
  2. A video presentation, not exceeding 5 minutes, explaining how your team approached solving the problem, including details of the model, and demonstration of the notebook. The video should be uploaded to YouTube, and the public link shared with the organizers via Devpost.

Hackathon Sponsors

Prizes

2 non-cash prizes
Apple iPad
1 winner

Every member of winning team will each get an Apple iPad

Amazon Echo
1 winner

Every member of runner up team will each get an Amazon Echo Studio

Devpost Achievements

Submitting to this hackathon could earn you:

Judges

Hugo Sancen

Hugo Sancen
ESG Data Product Owner/Sustainable Fitch

Derek Ferguson

Derek Ferguson
CSO, FitchGroup

Judging Criteria

  • Presentation Video
    A 3-5 minute presentation video needs to be included.
  • Code Repository
    All the codebase needs to be committed to a github repository and needs to be submitted as part of the solutions.

Questions? Email the hackathon manager

Tell your friends

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.