Voice AI Developer

BSG is a Communication Platform as a Service (CPaaS) that helps businesses connect with customers across multiple channels through flexible, cloud-based communication solutions.

We are looking for a strong backend/software engineer who can design and build scalable production Voice AI systems, with a focus on distributed systems, real-time processing, asynchronous architecture, streaming, and low-latency voice interactions.

Current tech stack: Python, FastAPI, Django, RabbitMQ, Memcached, JavaScript, REST, WebSocket, AI/Voice APIs.
Your Responsibilities

- Design and develop production-grade Voice AI systems and backend services.

- Build the full voice pipeline: Audio STT/ASR LLM TTS Audio .

- Integrate external and self-hosted AI/Voice models and APIs .

- Build real-time streaming services for bidirectional voice conversations.

- Work with streaming STT, LLM and TTS and optimize end-to-end latency and time-to-first-audio.

- Implement conversation state, context management and interruption/barge-in handling .

- Integrate Voice AI with telephony, telecom and business systems .

- Develop APIs and asynchronous, event-driven services using message brokers.

- Ensure high performance, reliability, scalability and observability of production systems.

- Monitor and analyze latency, recognition quality, voice quality, errors and conversation stability.

- Design solutions capable of supporting high volumes of concurrent calls .

- Participate in architecture decisions, code reviews and technical improvements across projects.

What We Expect

- 5+ years of commercial backend/software development experience.

- Strong proficiency in Python, Go, Java, Node.js/TypeScript, C# or another modern backend language .

- Solid understanding of OOP, APIs, concurrency and asynchronous programming .

- Experience building and maintaining production-grade backend services .

- Strong understanding of distributed systems and event-driven architecture .

- Experience with REST APIs, WebSockets and real-time/streaming systems .

- Experience with message brokers such as RabbitMQ or Redis.

- Ability to work with a new programming language and technology stack; Python is a strong advantage as it is the primary backend language in the current stack .

- Experience integrating external APIs, ideally including AI/LLM/Voice APIs .

- Understanding of Docker, CI/CD, monitoring and logging .

- Ability to design scalable, reliable and observable systems .

Nice to Have

- Voice AI / conversational AI / Speech-to-Speech

- STT / ASR / TTS / LLM

-
SIP, VoIP, WebRTC, RTP or SIP trunking

- Asterisk, FreeSWITCH or similar telephony platforms

- Real-time audio processing and codecs

- Low-latency real-time systems

- Call-center or telecom infrastructure

- Experience with self-hosted AI models

What We Offer

- Fully remote work with a flexible schedule.

- Competitive salary and 24 days of paid vacation .

- A technically strong team open to innovation and collaboration.

- Work on AI and communication products for the global market.

- Professional growth and participation in complex, technically challenging projects.

Originally posted on Himalayas