SoftwarePlacement Group
AI Recruiting and Sourcing Platform
SoftwarePlacement Group is a recruiting firm. We built an AI platform that helps its recruiters discover, enrich, score and match candidates to open jobs. It connects OpenAI’s models to the firm’s applicant tracking systems, so the ranking happens inside the workflow recruiters already have.
The platform parses resumes into structured data, scores each candidate against a job description using embeddings, and syncs candidates, job orders, submissions and status with PCRecruiter and Top Echelon in both directions.
The route
How this project climbed.
- 01Origin Point
The goal: help recruiters find the right candidate faster, including the ones already in the database.
- 02Blueprint Base Camp
A document model in MongoDB for resumes, jobs, enrichment data, scores and interaction history.
- 04Wireframe Switchbacks
One API layer designed to normalize two applicant tracking systems.
- 05Commit Ridge
Resume parsing and scoring with OpenAI APIs and embeddings, queue-based ingestion and services on AWS.
- 06Staging Camp
Logging, audit trails and PII safeguards around candidate data.
- 08Go-Live Summit
Ranked shortlists, search and two-way sync with PCRecruiter and Top Echelon.
Task
SoftwarePlacement Group wanted its recruiters to find the right candidate for a job faster, including the ones already in the database under a different job title. The task was to parse resumes into structured records, score candidates against job descriptions by meaning, enrich records with the attributes recruiters search by, and keep everything in sync with the firm's applicant tracking systems, with candidate data handled securely.
SoftwarePlacement Group is a recruiting firm. We built an AI platform that helps its recruiters discover, enrich, score and match candidates to open jobs. It connects OpenAI’s models to the firm’s applicant tracking systems, so the ranking happens inside the workflow recruiters already have.
The platform parses resumes into structured data, scores each candidate against a job description using embeddings, and syncs candidates, job orders, submissions and status with PCRecruiter and Top Echelon in both directions.
Parsing Resumes Into Structured Data
Resumes arrive in every layout. The platform reads each one with OpenAI’s models and converts it to structured JSON, then stores it in MongoDB alongside job descriptions, enrichment data, scoring outputs and interaction history.
Matching by Meaning
Each job and each candidate gets an embedding, a numeric representation of meaning. The platform measures how close the two are and reads that alongside structured attributes, which surfaces candidates a keyword search would miss.
Enrichment and Re-Scoring
Automated workflows append structured attributes such as skills, tenure, seniority and company history, and score a candidate’s fit again when the record changes.
One Layer Across Two ATS Platforms
Direct integrations with the PCRecruiter and Top Echelon APIs sync candidates, job orders, submissions and status updates in both directions. A unified API layer normalizes the data, so ranking, search and workflow automation work across both systems.
Built for Volume
Ingestion runs asynchronously through queues, so a large batch of resumes is processed in the background without slowing recruiters down.
AWS and Safeguards
- aAmazon EC2Compute and service hosting.
- bAmazon S3Document ingestion, storage and pipeline staging.
- cAmazon SESTransactional email and automated outreach.
- dAmazon AuroraRelational workflows beside MongoDB.
- eIAM and security groupsLeast-privilege access for every service.
- fCloudWatchService health monitoring and alerting.
Logging, audit trails and PII safeguards cover how candidate information is stored and handled.
Platform at a glance
Ranking that lives inside the recruiter’s workflow.
Back end
- PHP
- Python
- MongoDB
- Amazon Aurora
Integrations
- PCRecruiter API
- Top Echelon API
- OpenAI API
Operations
- Queues
- Audit trails
- IAM
- CloudWatch