Laiers.ai

Laiers.ai Logo

Inspiration

We were frustrated by how little traditional hiring tells you about the person behind the résumé. Candidates can appear ideal on paper but still be a poor fit in practice—especially when the real differentiators today are soft skills like emotional intelligence, adaptability, and leadership.

At the same time, we realized that much of the hiring process—job description creation, resume screening, even initial assessments—can be automated. What’s missing is personalization and insight. We built Laiers.ai to bridge that gap: to help companies understand who they're hiring, not just what they’ve done.

What it does

Laiers.ai is an AI-powered job matching platform that helps companies hire based on soft skills rather than just credentials. It includes:

AI-guided job creation customized to company needs
Automatic generation of soft skill assessments tailored to each role
Candidate evaluations using large language models trained to surface strengths, red flags, and fit
A company portal to manage opportunities and view ranked applicants
A talent interface for discovering jobs, applying, and tracking submissions
A chat-based UI that guides both companies and candidates through each step

How we built it

We built Laiers.ai as a full-stack web application using:

FastAPI for the backend and HTMX + Jinja2 for the frontend
Google’s Agent Development Kit (ADK) to create a multi-agent system for job creation and candidate assessment
Firebase Auth and Firestore for secure authentication and real-time data
Vertex AI and Gemini models for soft skill analysis and candidate ranking
Docker, Cloud Run, and Secret Manager for production deployment
A custom-designed interface that maintains a clean, professional user experience across user types

Challenges we ran into

Neither of us are developers by trade, so building Laiers.ai meant learning everything from the ground up—FastAPI, HTMX, Firebase, Google Cloud, ADK, and LLMs.

We had to figure out how to structure agent flows, manage asynchronous communication, debug deployment issues, and ensure that the AI was interpreting survey responses in meaningful ways. Diving into GenAI with limited background meant a steep learning curve, but also a big payoff.

Accomplishments that we're proud of

We created a fully functional AI hiring assistant without prior experience in machine learning or software engineering. We designed an intuitive UI, built real-time AI assessments, and deployed a production-grade system with authentication, data persistence, and a multi-agent backend.

Most importantly, we delivered a product that solves a real-world problem in a novel way.

What we learned

How to build and deploy an AI-powered web app from scratch
How to use Google’s ADK and Vertex AI to drive real-time assessments
How to design for both companies and job seekers in a single unified experience
How to prompt and structure LLMs for non-trivial tasks like ranking candidates or suggesting interview questions
The importance of clear user flow, dynamic personalization, and error handling in building trust with users