As AI continues to reshape the recruitment landscape, companies are increasingly moving beyond keyword-based search tools and toward intelligent, context-aware solutions. Client, a job marketplace platform, partnered with Addepto to take this leap and design and deploy an AI-driven candidate matching engine that delivers far greater relevance and efficiency in talent discovery.
Meet Our Client
The company is a multi-industry job marketplace that connects employers with job seekers through a digital platform. Users can create profiles, upload CVs, and apply for jobs, while companies can source talent through manual search or AI-enhanced recommendations.
Case Study Shortcut
Challenge
Accurately interpret and extract relevant information from free-text PDF CVs to identify qualified candidates
Surface strong candidate profiles that had incomplete or missing structured field data
Consistently apply recruiter expertise and business logic when ranking and selecting candidates for specific roles
Goal
The objective was to design a modular, AI-powered matching engine that could connect with Client’s existing core platform via API and:
Deliver accurate candidate-job matching based on context, not just keywords
Interpret unstructured data to expand the accessible talent pool
Be scalable to other job categories beyond finance
Outcome
The solution transformed the Client’s recruitment engine. It enabled nuanced, context-aware candidate matching, prioritized recruiter-relevant results, and vastly improved outcomes for roles with incomplete or unstructured profiles.
Before
Rule-based keyword matching
Structured fields only
Fixed rules
After
LLM-driven semantic understanding
Unstructured PDF CV analysis
Dynamic scoring with vector embeddings
Integrate those solutions in your company
Contact below and let us design and integrate solutions tailored to your business needs
Addepto developed a standalone microservice that combines structured filtering, LLM-based contextual understanding, and vector similarity scoring to intelligently match candidates with job requirements.
Structured Filtering Engine
The system applies configurable business rules for location, remote work eligibility, and other criteria using both profile fields and extracted CV data.
Contextual AI Analysis
Large language models interpret candidate experience, seniority levels, and domain expertise directly from PDF documents to understand qualifications beyond keywords.
Vector Similarity Scoring
Advanced algorithms calculate relevance scores by measuring semantic similarity between job requirements and candidate profiles in vector space.
Safe Development Environment
The team operated entirely within a cloned, non-production database environment to ensure zero disruption to live recruitment operations.
Industry-Agnostic Architecture
The solution was engineered for scalability and future deployment across all job categories within the client's ecosystem.
Timeline
Solution Design | 1 week
Prototyping | 2 weeks
Developement and Testing | 4 weeks
Technology
Docker (containers)
FastAPI
AWS
Python
Our Team Expert Opinion
This project wasn't just about building an algorithm - it was about modeling human judgment. We worked closely with Client to “translate” recruiter logic into measurable AI behavior, delivering a matching engine that truly understands relevance in hiring.
Filip MomotJunior Data Scientist at Addepto
Take the next step
Schedule an intro call to get know each other better and understand the way we work
Addepto, a fast-paced, growing company focused on innovations in AI-related and data-oriented areas, supports digital transformation at companies working on electronics manufacturing services.
Here you can learn more about the technologies used in this project:
Transform Engineering Chaos into Strategic Clarity
Discover how AI turns CAD files, ERP data, and planning exports into structured knowledge graphs-ready for queries in engineering and digital twin operations.