HumanBit Logo

Conversational AI Engineer | Scrabble

full-time
Posted on January 21, 2026

Job Description

Conversational AI Engineer

Job Summary

We are building voice-first AI companions for children aged 3–8 years. This role sits at the intersection of Large Language Models (LLMs), speech, linguistics, and product behavior. As a Conversational AI Engineer, you will own the AI's capabilities in communication, listening, memory, and behavior—ensuring interactions are safe, natural, and playful. This position does not involve chatbot responsibilities; instead, you will design real-time conversational systems with strict guardrails, latency constraints, and kid-safe behaviors.

Responsibilities

  • Design and improve multi-turn conversational flows tailored for children and parents.
  • Work on prompt engineering, system prompts, guardrails, and policies.
  • Tune AI behavior for:
    • Language switching (Hindi / English / Hinglish).
    • Age-appropriate responses.
    • Safety and content moderation.
  • Analyze real conversations to detect:
    • Hallucinations.
    • Broken flows.
    • Speech-to-Text (STT) / Text-to-Speech (TTS) issues.
  • Collaborate with backend and voice infrastructure teams to enhance latency and quality.
  • Build internal tools to review, score, and improve conversations.
  • Experiment with multiple LLMs and select the appropriate model for each use case.

Qualifications

  • 3–7 years of experience in Natural Language Processing (NLP), Conversational AI, or Applied Machine Learning (ML).
  • Strong understanding of:
    • LLM prompting and system instructions.
    • Multi-turn dialogue systems.
    • Intent, context, and memory handling.
  • Experience working with real user conversations.
  • Comfortable reading logs, transcripts, and metrics.
  • Strong written communication skills (very important).

Preferred Skills

  • Experience with voice pipelines (STT → LLM → TTS).
  • Experience with chatbots, assistants, copilots, or voice agents.
  • Familiarity with OpenAI, Gemini, Anthropic, and open-source LLMs.
  • Experience designing safe AI systems, especially in kids' products, healthcare, or education.
  • Exposure to nuances of Indian languages is a significant plus.

Experience

  • 3–7 years of relevant experience in Conversational AI or a related field.

Environment

  • The position is based in Gurgaon,
Powered by
HumanBit Logo