Engineering

AI Engineer (Voice and Language)

India, remote · Full time · Posted Oct 17, 2026

Make the AI features feel effortless: voicemail summaries, live call translation, call notes and caller screening. You will own prompts, evaluation, latency and cost across speech recognition, Claude and text-to-speech.

What you will do

  • Design and evaluate prompts and output schemas for summaries, translations and action items, with test sets in English, Arabic, Urdu and Hindi.
  • Cut end-to-end latency on live translation: transcription segmentation, model calls and speech synthesis.
  • Build evaluation harnesses so quality is measured, not felt, and track cost per minute by feature.
  • Add new AI capabilities with the product team, such as spoken-language detection and smarter screening.
  • Harden the pipeline against prompt injection and abuse.

What we are looking for

  • Three or more years building with language or speech models in production, including evaluation work.
  • Strong Python or TypeScript; comfort with streaming audio and webhooks.
  • Fluency in English plus Hindi or Urdu; Arabic is a plus.
  • Based in India, with four hours of daily overlap with US Eastern time.

Hiring process

  • Apply with your resume and a short note on an evaluation you ran.
  • A 30-minute conversation.
  • A paid take-home.
  • A final conversation, then an offer.

Pay and location

INR 2,00,000 to 3,50,000 per month

INR 2,00,000 to 3,50,000 per month depending on experience, paid monthly.

India, remote. Equipment allowance and paid time off included.

Apply

Five fields and your resume. We reply to everyone within a week.

Your details are used only for this hiring process and deleted on request.

FreeCallNow is an equal opportunity employer. Questions about the role: support@freecallnow.com.