Build a Custom System to Prequalify Recruiting Candidates
Choose a partner who delivers full source code, not a per-seat SaaS subscription. Select the engineer who builds your system directly, not a firm that outsources the work.
The right approach depends on your applicant volume and existing Applicant Tracking System (ATS). A system for a firm processing 500 applications a month into Greenhouse requires a different setup than one handling 100 applicants arriving via email and spreadsheets. The complexity lies in the custom logic required to properly screen for technical roles or specific certifications.
We built a pre-qualification system for a 12-person recruiting firm that was spending 15 minutes per applicant on initial phone screens. The new system uses a 90-second automated call that feeds structured notes directly into their ATS. The entire build was live in three weeks, eliminating thousands of hours of manual work per year.
What Problem Does This Solve?
Most recruiting teams first try off-the-shelf AI interview platforms. These tools seem attractive because they offer a quick setup, but their business models fail small businesses. They typically charge per seat or per interview, meaning a 10-person firm could pay over $1,500 per month for a single function, with costs ballooning during a hiring surge.
A more critical failure is their rigid logic. These platforms let you customize questions but cannot dynamically branch the conversation. For example, a technical recruiting firm needs to screen software engineers. Their platform asks, "Do you have experience with Python?" It can score a "yes" answer, but it cannot ask a follow-up like, "Describe a time you used an async library like httpx." This means a junior developer who took one online course gets the same score as a senior engineer, creating hours of wasted effort for human recruiters who must re-qualify everyone.
These platforms are designed for high-volume, generic screens, not for the nuanced qualification that small, specialized firms depend on. They cannot distinguish shallow keyword matches from deep expertise, which is the entire point of a technical screen. The result is a noisy, unreliable signal that damages recruiter productivity.
How Does It Work?
We begin by mapping your existing screening questions into a decision tree and connecting directly to your applicant source, whether it's an ATS like Lever or a dedicated inbox. We use the Claude API for its conversational ability and strong instruction-following to design the interview flow. The complete script and its branching logic are stored in a Supabase database, allowing you to edit questions without a code change.
The core of the system is a Python application built with FastAPI. When a new candidate applies, a webhook triggers an AWS Lambda function, which initiates the automated call in under 500ms. We use a dedicated voice API to handle real-time transcription and speech synthesis. The system's logic is a state machine that pulls the next question from Supabase based on the transcript of the candidate's previous answer, allowing for truly dynamic conversations.
After a call completes, typically in 90 to 120 seconds, the full transcript is sent back to the Claude API. A specific prompt extracts key information: years of experience, salary expectations, specific skill mentions, and potential red flags, outputting a clean JSON object. This data is then used to populate custom fields in your ATS or a new row in a Google Sheet. The entire post-call processing pipeline runs in under 8 seconds.
We deliver the complete source code to your private GitHub repository and deploy the system on your own cloud infrastructure. We configure structured logging using structlog and set up alerts that fire to Slack if the API error rate exceeds 2% or if transcription accuracy falters. The monthly hosting cost for processing up to 1,000 candidates is typically under $75.
What Are the Key Benefits?
First Candidates Screened in 15 Days
Go from kickoff to a live system screening your actual applicants in three weeks. Stop manual phone screens this month, not next quarter.
Pay for Usage, Not for Seats
A one-time build cost and low monthly cloud fees based on call volume. Your costs scale with your hiring needs, not your headcount.
You Own the Code and the Logic
Receive the complete Python source code in your GitHub. You are never locked into a vendor and can have any developer extend the system.
Alerts for Problems, Not Reports for Vanity
We configure monitoring that alerts your team in Slack if a key process fails. No more discovering a problem when a candidate complains.
Writes Directly Into Your ATS
Structured notes and pass/fail scores are posted directly to candidate profiles in Greenhouse, Lever, or Ashby. No copy-pasting from transcripts.
What Does the Process Look Like?
Workflow Mapping (Week 1)
You provide your current screening questions and grant read-only access to your ATS. We deliver a complete process diagram showing the new automated flow and the decision logic.
Core System Build (Week 2)
We write the FastAPI application and the Claude API integration logic. You receive access to a staging environment to test calls and review the generated candidate summaries.
Integration and Deployment (Week 3)
We connect the system to your live ATS and deploy it to your cloud infrastructure. We provide a runbook detailing the architecture and how to update screening questions.
Monitoring and Handoff (Weeks 4-6)
We monitor the first 100 live candidates, tune summary prompts, and adjust scoring logic. After this period, the system is fully handed over with an optional maintenance plan.
Frequently Asked Questions
- How much does a custom recruiting screener cost?
- The cost depends on the complexity of your screening logic and the number of systems we need to integrate. A simple system connected to a standard ATS is typically a 3-week build. A system requiring complex, branching logic or integration with a custom-built CRM will take longer. We provide a fixed-price quote after a 30-minute discovery call at cal.com/syntora/discover.
- What happens if a call drops or the AI misunderstands a candidate?
- The system is designed for graceful failure. If a call drops, the candidate is flagged for manual review in your ATS with a note explaining the issue. If the AI cannot parse an answer, it asks for clarification once. If it still fails, that question is marked as 'unanswered' in the summary, alerting your human recruiter to follow up during their review.
- How is this different from using a platform like HireVue?
- HireVue and similar platforms are comprehensive suites with video interviews, scheduling, and assessments, designed for large enterprises. Syntora builds a targeted tool that solves one problem: initial phone screening. You get a production-grade system you own completely, without paying for features you do not use or being locked into a multi-year SaaS contract.
- Will this create a bad experience for candidates?
- The introduction script is transparent, stating it's an automated screening call. The voice is natural, and response latency is very low. Candidates often prefer it for a first screen as it is quick and can be done anytime. It improves the experience by giving every applicant an immediate chance to be heard, rather than waiting weeks for a human screener to become available.
- How do we change screening questions for a new role?
- The questions are stored in a Supabase table, not hard-coded in the application. We provide a simple interface where you can edit, add, or remove questions for different job roles without needing a developer. The changes are live as soon as you save them. For changes to the underlying branching logic, a small support ticket is required.
- How is candidate data handled and stored for privacy?
- The entire system is deployed on your own cloud infrastructure (e.g., AWS). Candidate data and call transcripts never pass through Syntora's servers after the handoff. Data is encrypted in transit and at rest using standard cloud provider tools. You maintain full control and ownership over all personally identifiable information, helping you comply with GDPR and CCPA.
Related Solutions
Ready to Automate Your Small Business Operations?
Book a call to discuss how we can implement ai automation for your small business business.
Book a Call