CTRL-ALT-DELIGHT

Framing the Problem

https://www.canva.com/design/DAGrue4Lanw/x_R6TW2ktr_EP2DLXdM_cA/edit?utm_content=DAGrue4Lanw&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton

What is the problem you are trying to solve? Who does it affect?

For many underrepresented communities, access to meaningful job opportunities remains limited. Even as more roles and employers emerge, pathways to these positions are often scattered, poorly advertised, or hidden within exclusive networks—making discovery overwhelming and inaccessible.

Traditional job boards focus narrowly on skills and experience, offering little insight into whether companies value diversity, support flexible work, or uphold environmental responsibility. This lack of transparency hinders job seekers from making informed decisions and limits access to inclusive, values-aligned workplaces—while also preventing employers from effectively showcasing their commitment to equity and sustainability.

Idea Explanation

What is your idea? How does it fix the problem?

GreenFlag is a values-driven job discovery platform that helps job seekers—particularly those from underrepresented communities—identify inclusive, equitable, and environmentally responsible employers. Unlike traditional job boards that prioritize qualifications alone, GreenFlag highlights what truly matters: company culture, support policies, and sustainability commitments embedded in job descriptions.

By analyzing job postings through web scraping and AI-powered content analysis, GreenFlag surfaces key indicators such as flexible work arrangements, mental health benefits, inclusive hiring practices, and green initiatives.

The platform’s Key Values Search allows users to filter and prioritize opportunities based on personal values—whether that’s pay transparency, fertility support, or environmental impact.

GreenFlag not only empowers job seekers to make informed, values-aligned career decisions, but also supports employers in showcasing their commitment to diversity, inclusion, and sustainability. It bridges the gap between intent and access—transforming how job opportunities are found, evaluated, and shared.

Implementation

How do all the pieces fit together? Does your frontend make requests to your backend? Where does your database fit in?

Our solution leverages advanced web scraping technology combined with artificial intelligence to provide comprehensive job posting analysis. The application uses Nokogiri for HTML parsing and content extraction, enabling it to analyze any job posting URL and extract meaningful insights. The system employs a sophisticated keyword detection algorithm that searches for 17 different categories of inclusive language, ranging from fertility support and pay transparency to religious inclusion and carbon neutrality initiatives.

The application is built on a modern Rails 8.0.2 stack with Ruby 3.4.2, utilizing Hotwire and Stimulus for dynamic frontend interactions. The backend processes web content through multiple analysis layers: text extraction, word frequency analysis, table of contents generation, and inclusive terms detection. We integrate OpenAI's GPT-4.1-mini API to provide intelligent job summaries that highlight positive aspects, potential red flags, and concise overviews of each position. The system stores all analysed data in a SQLite database, allowing users to access historical analyses and track trends over time.

Our comprehensive CI/CD pipeline utilizes GitHub Actions to automatically run security scans with Brakeman, enforce code quality with RuboCop, execute comprehensive unit and system tests, and perform smoke tests before deployment, while Dependabot manages daily dependency updates to ensure the application remains secure and up-to-date.

Challenges

What did you struggle with? How did you overcome it?

Four of us had never touched Ruby-on-Rails so it was fun to pick that up. We had some issues getting it working on Windows, but through sharing our learnings and pair-programming our way through the errrors - we got the Rails Server running for all of us. As it took us longer to set up, we had to scale back on some of our vision to prioritise getting the core functionality working in our MVP.

Accomplishments

What did you learn? What did you accomplish?

• Learnt Ruby!

• Learnt how to integrate OpenAI calls within Ruby & display the content on the Ruby frontend

• Excalidraw is great for extremely quick wireframing

• MoSCoW prioritisation: Must-have, Should-have, Could-have

• Got user auth working, keyword job in job descriptions, openAI job summary feature + library of previous searches

Next Steps - Our Vision for GreenFlag

• When someone searches for a job, scrape that company's 'Careers' page to supplement the information from job description keyword search

• Potentially get information from review websites like Glassdoor to sense-check whether the company lives up to the policies highlighted in the job description

• Change the database to Neo4j so that overtime we could build a map of companies + job descriptions that match

• Harness the power of community with a feature to email companies that don't have certain keywords