Multilingual Subsidy Qualification Agent

An energy sector client needed to qualify homeowners for a subsidy program over the phone, screening for eligibility criteria before passing qualified leads onward.

1. 01
   Qualification script

   A voice agent asks a structured set of eligibility questions covering the homeowner's property and situation, built to match the subsidy program's specific criteria.

2. 02
   Cross lingual build and testing

   The agent's core logic and prompts were built in English, then debugged and validated through live conversations conducted in a different language, confirming the conversational flow held up across languages.

3. 03
   Outcome routing

   Based on the answers given, the system determines whether the caller qualifies and routes the outcome accordingly (qualified lead handoff vs. disqualified/closed).

Originally built for the energy sector, the same architecture applies to any team facing a similar volume, compliance, or coordination problem. The stack (ElevenLabs, Twilio, Multilingual prompt design) is stable, self hosted where it matters, and designed to be operated by a small team without a platform crew.

• multilingual voice AI agent
• AI eligibility screening
• voice AI for energy sector
• subsidy qualification automation

• ElevenLabsSpeech synthesis for the voice agent. Chosen for latency and voice quality on production calls.
• TwilioTelephony carrier. Handles inbound and outbound PSTN, SMS, and recording.
• Multilingual prompt designPart of the production stack for this build.

I would invest earlier in a proper eligibility rule engine rather than encoding rules in the agent prompt. It makes audits easier and lets non technical staff update criteria without a deploy.

Building and testing a voice AI agent across languages. A good piece on the practical challenges of multilingual conversational AI.