Generative AI Engineer

About YABX

Yabx is a global FinTech venture of Mahindra Comviva aimed at simplifying financial access to the 2 Bn+ under-banked people in the emerging markets of Africa, Asia, and Latin America. We use technology and analytics to reduce the cost of delivering financial services. In doing so, we partner with leading Telecom Operators, Banks, MFI's, Credit Bureaus, Mobile Financial Providers & handset vendors. Yabx's mission is to create world-class innovative products to improve the lives of these people.

Yabx was incubated by Mahindra Comviva and operates between The Hague, New Delhi, and Nairobi. We are a team of industry experts and entrepreneurs with global experience.

Work Location: Gurgaon

Total Experience: 2-5 years

Position: AI Engineer (LLM & Voice AI)

Job Overview

We are building next-generation AI systems powered by Large Language Models and Voice AI. Our focus is on creating real-time conversational systems, AI agents, and intelligent voice applications used in production environments.

We are looking for an AI Engineer who enjoys building real systems, not just models — someone who can design and deploy LLM-powered applications, voice assistants, and real-time AI pipelines.

This role involves working at the intersection of LLMs, real-time voice systems, telephony infrastructure, and conversational AI workflows.

What You Will Build:

• LLM-powered assistants and agents
• Voice AI applications (AI calling systems, voice assistants)
• Retrieval-Augmented Generation (RAG) pipelines
• Real-time conversational AI systems
• AI-powered workflows integrated with telephony systems

You will work on production AI systems used in real-world environments, not research prototypes.

Core Responsibilities

• Build and deploy LLM-powered applications including chatbots, assistants, and AI agents.
• Develop scalable APIs and services using Python, FastAPI, or Flask.
• Implement RAG pipelines using embeddings, vector search, and retrieval systems.
• Integrate Speech-to-Text (STT) and Text-to-Speech (TTS) technologies.
• Design memory and context management systems for conversational AI.
• Build real-time AI interaction systems using WebSockets or streaming APIs.
• Monitor and evaluate AI systems using prompt management and observability tools.

Required Skills

• Experience building LLM-based of conversational AI applications (chatbots, assistants, agents).
• Strong proficiency in Python.
• Experience with FastAPI or Flask for building backend services.
• Experience integrating Speech-to-Text and Text-to-Speech systems (Deepgram, ElevenLabs, etc.).
• Experience with RAG pipelines, embeddings, and vector search.
• Understanding of conversational AI workflows (memory, context, session handling).

Nice to Have

• Knowledge of SIP, RTP, VoIP, or WebRTC telephony systems.
• Experience with Langfuse, Phoenix, or other LLM observability tools.
• Experience with WebSockets (WSS) for real-time AI systems.