The past 3 years have been one of the most tumultuous years in our lifetime, in terms of global health upending by the pandemic, unstable geo-politics, and an extremely volatile economy. The US tech industry, alone, went through the most devastating layoffs in history, where more than 250,000 tech professionals lost their jobs.

We believe, at Yeonjii, that everyone deserves the best chance to land their dream job of their choice, regardless of any conditions. Our team is focused on delivering the best AI products to empower job seekers.

Problem

To empower job seekers, we have to alleviate the core problem of time being spent and quality of the application. We see 4 major components of the job application as being the resume, interviews, professional outreaches, and cover letters. Our first product, we chose to focus on the cover letter here is why.

Research + Data

To get to the core problem and the subtleties of the cover letter and how it fits with the overall recruiting process, we spoke to recruiters to get their insights on what they look for in a cover letter, and to assess if it does provide any value for job applicants

“Most cover letters are not read unless it’s unique, which means it needs to sound personal and have a hook” - primary research via interview
“Conciseness and readability is the second most important factor” - primary research via interview
The chances of giving preferences to an applicant are 60% more likely for a personalized cover letter - secondary research

From 20+ survey responses and 5 interviews, we found 75% of job seekers didn’t know how to write a proper cover letter and the majority of them didn’t see any value in making one. However, a striking data point caught our eye. 90% of tech professionals, still wrote cover letters. Spending an average of 10 hrs per week writing them, despite the fact, they believed it provided them little to no value

Additionally, job seekers in the tech space don’t necessarily find writing cover letters so easy. 77.3% of tech professionals, found writing a cover letter to be moderately to highly difficult.

Scale of 0-10, 10 being the most difficult. 0 being not difficult at all

Forms response chart. Question title: Please rate the level of difficulty you experience when writing cover letters.. Number of responses: 22 responses.

So why spend hours on a document, when no one sees the value in doing them? The reason is that in a highly competitive job market coupled with weak economic conditions, users believed that it optimized their chances of landing the 1st interview with a hiring manager.

Solution

Our generative AI-powered cover letter creator is able to generate a stand-out personalized cover letter in seconds. We focus on 3 key features to bring

Simple and intuitive UX/UI design: Hyper focused on user-centricity
AI-based text editing guidance feature: providing suggestions on how to improve the cover letter
AI-powered personalized career-mentor chatbot: allowing job seekers to be mentored and guided throughout their job application process

Design (UX/UI)

Yeonjii, inspired by the Korean word "Pyeonji" meaning "letter," and made distinctive with the creative omission of 'P', stands as a beacon of our commitment to user-centered design. Our innovative approach is deeply woven into every aspect of Yeonjii, from its name to its functionality.

The design phase progressed from initial wireframes—outlining the fundamental UX structure—to a phase of design exploration where we refined visual and interactive elements. This led to the creation of Yeonjii's design guidelines, which ensure visual and functional consistency across the platform.

At the heart of Yeonjii’s development has been a rigorous, data-driven process informed by direct user feedback. User interviews have been invaluable, providing us with the insights needed to shape a UI that not only meets but exceeds user expectations, reflected in a stellar 100% positive feedback rate. These interviews revealed a significant user demand for guided assistance within the platform, with 80% of users indicating the need for more support—a key insight that has been integral to our user-centered design focus.

Usability testing has played a critical role in refining Yeonjii's features to ensure they are user-focused:

The unanimous use of PDF resumes among our testers (100%) led to the seamless integration of a PDF resume upload feature, directly addressing user needs and streamlining the application process.
An average Net Promoter Score (NPS) of 7 showcases the platform’s ability to engage users and the likelihood that they would recommend Yeonjii to others.
The across-the-board willingness to pay for Yeonjii services underscores the platform's perceived value in aiding users on their job search journey.

This unwavering dedication to a user-centered design ethos ensures that Yeonjii isn’t just a tool but a trusted partner for job seekers, facilitating their pursuit of career opportunities with confidence and ease.

Implementation Details ‍

Hosted at https://yeonjii.netlify.com‍

What is your tech stack?

Flask, Python, React, Typescript, TailwindCSS, Vite.

Current Design:

The design of our application is user centric driven. We chose to build the application to view on a desktop/laptop browser. However, we do foresee that taking our solution and making it more accessible via web extension in the future.

Technical challenges

What was the hardest part of development?

Frontend: The most challenging part of the frontend was implementing a modal/carousel for the user to input their information while ensuring error handling

This also entailed sending the data to the proper routes and handling errors.

Backend:The most challenging part of the backend was prompting OpenAI, specifically ChatGPT 3.5 Turbo API. The difficulty in this task was taking user data from the frontend and providing a unique cover letter

We used ChatGPT 3.5 Turbo API because it was cost-effective for our solution. However, we believe implementing ChatGPT 4.0 will provide significant space to scale our solution in the future.

Trade Offs: Could we have built our product using one type of language, the simple answer is yes. But we chose to use two different languages because of the complexity of OpenAI on the backend, which is written in Python.

Key takeaways

Adaptability:

We had to solve many issues to ensure our product was always usable.

We conducted 6 user testings, in-person interviews. We were able to validate our hypothesis which some had proven that we were on the right track, and a few feedbacks that we did not expect.

Simplifying personal story input and giving specific instructions on what to write so users don’t get stuck
Prompt engineering so the cover letter doesn’t sound generic and impersonal

Technical Proficiency:

Learning the depths of Git, Typescript, Redis and prompting AI was an experience that will help us in the future.

Team Coordination:

Being open and honest with all team members helps each other know what could be accomplished versus what to put aside for another date.

Long-term goals/ vision

“We will be continuing with our product and continue to refine our cover letter product . Our vision is to become a one-stop shop for job seekers where we provide the best products and services to manage your job application from start to finish.

‍

Learnings

Product Manager Learnings:

Phil Kim

Strong research sets the foundation: Do the research even if it takes the time. Strong research = insightful/actionable feedback = impactful features

Clarity in vision: Be clear with your vision. There are endless amounts of ideas and product decisions. Be focused on a singular vision when ideas go rampant

Stay disciplined and focused: some PMs may be overly ambitious to make the best product they can, which can lead to adding more, but simplicity - prioritizing on what has the highest impact that brings the most value to users.

Designer Learnings:

Jose Quezada

Non-Linear Design Insights: My experience taught me about the complexities of non-linear design processes, enabling me to better align creative efforts with diverse team requirements and focus on user-centric design principles.

Collaboration and Communication: I developed a deeper understanding of the need for continuous and clear communication within teams to maintain alignment and facilitate progress.

Interdisciplinary Communication Skills: Engaging with developers provided valuable lessons in effectively conveying design concepts, enhancing my ability to bridge the gap between design and technical perspectives.

Developer Learnings:

Austin Beveridge

Adaptability

Adaptability was vital in navigating a dynamic project. Initially, using traditional libraries required a swift transition to modern frameworks based on evolving requirements. This experience enhanced problem-solving skills and highlighted the importance of adapting to the product manager's direction, product designers' visions, and backend developers' logic. Making real-time adjustments proved crucial, demonstrating adaptability beyond technology, and remains essential for success in the ever-evolving front-end development landscape.

Collaboration

Collaboration serves as the linchpin of success.- engaging seamlessly with the backend developer, product designer, and product manager while sharing a collective ownership of the project. Clear communication is the cornerstone of their collaborative efforts, ensuring a unified understanding of project objectives, timelines, and expectations by appreciating each team member's unique perspectives and skill sets, fostering a harmonious flow of ideas that enables them to collectively address challenges and craft an innovative frontend experience.

Prioritization

Prioritization is the cornerstone of success, emphasizing the strategic alignment of logic and functions over design aesthetics. Collaborating with backend developers, product designers, and product managers, the team is responsible for determining the order of tasks. Clear communication ensures tasks are prioritized with a shared understanding of the project goals, timelines, and expectations. This approach fosters a seamless flow of work, allowing the team to grasp and address critical aspects efficiently.

Developers Learnings:

Martin Ruthaivilavan

&

Team Collaboration and Skill Enhancement:

Enhanced skills in collaborative work, particularly in backend architecture, keeping frontend integration in mind. Coordinated closely with frontend developers to seamlessly connect both systems with each new feature. Regularly engaged with the PM and PD to merge user research into technically feasible features, finding this integration process highly stimulating.

Technical Feasibility: Developed a deep understanding of the Product Manager's (PM) and Product Designer's (PD) requirements, quickly assessing the feasibility of these within specific timeframes. Focused on setting realistic expectations and committing to thorough research prior to coding, a process that became more streamlined with the emergence of new features or ideas.

Agile Development: Gained initial exposure to AGILE development principles and methodologies. Adapted to rapidly changing project requirements, efficiently implementing new features like AI integration and cloud storage management. This role highlighted the significance of adaptability, iterative testing, and continuous refinement in backend development.

Full Team Learning

Over an intensive eight-week period, our diverse team, consisting of a product manager, a product designer, and two engineers, developed an AI-powered cover letter generator. One key takeaway from this experience was the importance of cross-functional collaboration; regular communication and brainstorming sessions were vital in aligning our diverse expertise towards a common goal.

We learned the value of iterative design and agile development, as rapid prototyping and user feedback loops significantly enhanced the functionality and user experience of our product. This project also highlighted the critical role of AI in automating and personalizing complex tasks, as our tool successfully streamlined the cover letter writing process, making it more efficient and tailored to individual user needs

Reiterations and trade-offs are essential: Your first idea of the product never comes out as initially thought. There are multiple trade-offs that needs to be made to ultimately provide maximum value to users