Build a Custom Voice AI for Applicant Questions
A voice AI agent can answer common applicant questions over the phone 24/7. It integrates with your Applicant Tracking System (ATS) to provide real-time status updates and pre-screen candidates.
Syntora offers a tailored service for integrating voice AI to handle inbound applicant queries. Our approach leverages a custom FastAPI application with Claude API and ATS integrations to provide 24/7 automated phone support, ensuring efficient and honest candidate communication.
Syntora's approach to a voice AI assistant for applicant queries involves connecting a large language model to your existing ATS and a telephony provider. The complexity and timeline for such a system depend heavily on the number of unique questions the AI needs to handle, the depth of required ATS integration, and the specific APIs available. A system designed only for static information retrieval is generally a faster build than one requiring real-time candidate data reads and writes. Syntora brings experience building similar document processing pipelines using Claude API for financial documents, and the same architectural patterns apply to structuring knowledge bases for applicant-related information.
What Problem Does This Solve?
Most small businesses start with a standard phone tree (IVR) from their VoIP provider. These systems fail because they are not conversational. An applicant asking, "Did you get my application for the forklift operator job?" cannot be answered by a "Press 1 for hours" menu. Callers inevitably press '0', defeating the purpose of the system.
A dedicated contact center platform like Talkdesk offers AI features, but they are designed for 100-plus agent teams and carry a per-seat monthly cost that is unfeasible for a small recruiting team. The setup and maintenance also require a dedicated IT specialist, which most small businesses lack. The alternative, trying to connect a web-based chatbot like Drift to a phone line, results in a frustrating experience with 2-3 second delays between speaking and hearing a response, making a natural conversation impossible.
This leaves the office manager or a junior recruiter spending hours each day answering the same repetitive questions: "Are you still hiring?", "What's the status of my application?", "What are the benefits?". This manual effort prevents them from focusing on sourcing and interviewing qualified candidates. The result is significant: a team spending 10 hours per week on repetitive phone calls can typically reduce that to under two hours with a well-built voice agent.
How Would Syntora Approach This?
Syntora's approach to integrating voice AI for inbound applicant queries begins with a discovery phase. We would identify the 20-30 most frequent questions your applicants ask and map these to relevant data fields within your ATS, such as Lever or Greenhouse. This process informs the creation of a structured knowledge base, which would be stored in a Supabase Postgres database, allowing the AI to query for static information like office locations or benefits summaries.
For telephony, the system would utilize Twilio to manage the phone number and handle the incoming call stream. When a person calls, their speech is converted to text via Twilio's capabilities and sent to a custom FastAPI application. This application would then send the transcribed query and conversation history to the Claude API, which generates a natural language response. The entire process is designed for low latency, aiming for a swift conversational turn.
The AI's text response would be converted back to audio using AWS Polly, a text-to-speech service known for its low latency. This audio would then be streamed back to the caller through Twilio. For dynamic queries that require real-time data, such as application status checks, the AI agent would employ a tool-calling function. This function would invoke an AWS Lambda function to securely query your ATS API. We have experience building robust API integrations and ensuring data security in sensitive contexts.
The core FastAPI service would be deployed as a Docker container on AWS App Runner, providing a managed and highly available environment. The system would log every call transcript and AI response to your Supabase instance, facilitating review and continuous tuning. We would implement structured logging with `structlog` to monitor performance, configuring alerts to a shared Slack channel if API latency from the ATS exceeds defined thresholds or error rates spike. Typical hosting costs for this architecture are generally under $50/month. A typical build timeline for a system of this complexity, from discovery to initial deployment, would be approximately 6-10 weeks, requiring client involvement for ATS access, knowledge base content, and testing.
What Are the Key Benefits?
Launch in 4 Weeks, Not 6 Months
From our initial discovery call to a live, production-ready voice agent in 20 business days. Your team gets relief from repetitive calls immediately, not after a quarter-long enterprise rollout.
Fixed Build Cost, Not Per-Agent SaaS
We deliver the system for a one-time project cost. After launch, you only pay for low-cost cloud hosting, with no recurring per-user or per-call subscription fees.
You Get the GitHub Repo and the Data
The complete Python source code is delivered to your GitHub account. All call logs and transcripts are stored in your own database, not ours. You have full ownership of the system.
Alerts on Failure, Not Silence
We build monitoring directly into the system. You get a Slack alert if your ATS API is down or if response times are slow, so you are never in the dark about system health.
Connects Directly to Your ATS
The agent integrates with the tools you already use. We have built direct API connections for Greenhouse, Lever, and Ashby, providing applicants with real-time, accurate information.
What Does the Process Look Like?
Week 1: Scoping and ATS Access
You provide a list of common applicant questions and grant read-only API access to your ATS. We deliver a finalized scope document outlining every question the agent will be trained to answer.
Week 2: Agent Development
We build the core conversational logic using FastAPI and the Claude API. You receive a private phone number to test the agent's ability to answer static questions from its knowledge base.
Week 3: ATS Integration
We connect the agent to your live ATS data via a secure AWS Lambda function. You receive access to test dynamic, real-time queries like checking an application's status.
Week 4: Launch and Handoff
We port your existing recruiting number or provision a new one and go live. After a 30-day monitoring period, you receive a technical runbook and a performance report detailing call volume and resolution rates.
Frequently Asked Questions
- How much does a custom voice agent cost?
- The cost depends on three main factors: the number of unique questions the agent must handle, the complexity of the ATS integration, and whether it needs to perform actions like scheduling. A read-only agent that answers 20 questions is simpler than one that pre-screens candidates and writes notes back to your ATS. We scope every project individually after a discovery call.
- What happens if the AI gives a wrong or nonsensical answer?
- The system has a fallback to transfer the call to a human if it cannot answer a question confidently. During the 30-day post-launch period, we review all call transcripts to identify and correct any incorrect responses by refining the AI's prompts. This tuning process dramatically reduces errors within the first few weeks of live operation.
- How is this different from a chatbot from Intercom or Drift?
- Web chatbots are designed for asynchronous text interaction. Our system is built for synchronous, low-latency voice conversations. We use specialized speech-to-text and text-to-speech engines like AWS Polly and structure the AI's logic for the turn-based nature of a phone call. This avoids the awkward delays and misunderstandings common when a text-based bot is retrofitted for voice.
- Can the agent handle different languages or accents?
- Yes. We use AWS Transcribe for speech-to-text, which supports over 100 language variants. We can configure the agent to operate entirely in another language, such as Spanish, or to better understand specific regional accents. This is defined during the scoping phase to ensure high accuracy for your specific applicant pool.
- Does it just answer questions, or can it take action?
- The agent can perform actions by writing data back to your ATS. A common use case is pre-screening. The agent asks a candidate required questions, like "Do you have a valid commercial driver's license?", and saves their yes or no answer as a note in their applicant profile. This automates the initial qualification step for high-volume roles.
- Do we need to get a new phone number?
- You have two options. We can instantly provision a new local or toll-free number for the agent through Twilio. Alternatively, we can port your existing recruiting phone number to our system. The porting process is managed with the carriers and typically takes 1-2 weeks, during which your existing service is not interrupted.
Ready to Automate Your Professional Services Operations?
Book a call to discuss how we can implement ai automation for your professional services business.
Book a Call