Back End Developer

Recrew AI; Bengaluru; On-site; INR 1,000,000 - 2,000,000; Full time

Benefits offered by this job

Opportunity to build impactful voice AI infrastructure

Access to next-gen GPU clusters

Competitive compensation and benefits

Job summary

Recrew AI in Bengaluru is seeking a Backend & Platform Engineer committed to developing high-concurrency voice streaming systems. The candidate will architect real-time streaming infrastructures that ensure high availability and low latency.

The role emphasizes building and integrating WebSockets, gRPC, and RTP/WebRTC protocols for seamless voice sessions. An ability to manage 1M+ connections and experience with Go or Rust is essential. The position promises competitive compensation and a unique opportunity to impact millions of users in India.

Qualifications

• 4–9 years of backend engineering experience building high‑concurrency, distributed production systems.
• Hands‑on experience with Go or Rust for performance‑critical backend services.
• Proven experience building real‑time streaming systems using WebSockets and gRPC at scale.

Responsibilities

• Architect and own the real-time streaming backend handling WebSockets and gRPC.
• Build audio chunk orchestration pipelines with ultra-low latency.
• Design and implement seamless state management across distributed voice streams.

Skills

Backend engineering experience

Experience with Go or Rust

Real-time streaming systems

Knowledge of RTP/WebRTC

Distributed caching using Redis

Architecting systems for high concurrency

Container orchestration using Kubernetes

Job description

Type: Full-time

Industry: Artificial Intelligence, Telecommunications, Voice AI, Real-Time Communications

About Company

The company is the dedicated AI research arm of a large-scale national intelligence program. It is building foundational AI for India at population scale.

Focus areas include Speech-to-Text, Text-to-Speech, Real-Time Conversational AI, and Multilingual Foundation Models. The company invests across the full AI value chain — from next-gen data centers to consumer AI platforms and enterprise enablement.

It partners with global AI leaders including OpenAI, Anthropic, Google, and Meta. The mission: make human-machine communication feel as natural as speaking to another person.

Position Overview

We're looking for a Backend & Platform Engineer who builds systems that handle millions of concurrent voice streams without breaking a sweat. You will architect the real-time streaming backbone — WebSockets, gRPC, audio chunk orchestration — that connects client-side edge to our core AI inference nodes. High availability isn't a goal here; it's the default setting.

Role & Responsibilities

• Architect and own the real-time streaming backend handling WebSockets and gRPC for live voice sessions
• Build audio chunk orchestration pipelines with ultra-low latency and zero data loss guarantees
• Design and implement seamless state management across distributed, concurrent voice stream sessions
• Engineer low-overhead routing layers between client-side edge nodes and core AI inference clusters
• Integrate RTP/WebRTC protocols into the platform for real-time audio transport and session handling
• Implement distributed caching patterns to minimize latency across millions of simultaneous connections
• Drive platform reliability, observability, and incident response for mission‑critical voice infrastructure

Must Have Criteria

• 4–9 years of backend engineering experience building high‑concurrency, distributed production systems
• Hands‑on experience with Go or Rust for performance‑critical backend services (not just familiarity)
• Proven experience building real‑time streaming systems using WebSockets and gRPC at scale
• Working knowledge of RTP/WebRTC protocols and audio codec handling (Opus, G.711, or similar)
• Experience implementing distributed caching using Redis for low‑latency state management
• Demonstrated ability to architect systems handling 1M+ concurrent connections or equivalent scale
• Experience with container orchestration using Kubernetes for deploying high‑availability services

Nice to Have

• Experience building voice or audio infrastructure for a conversational AI or VoIP product
• Prior work at a real‑time communications company (Twilio, Agora, Daily.ai, or similar)
• Familiarity with GPU inference serving and optimizing the edge‑to‑inference handoff latency
• Open‑source contributions to networking, audio processing, or distributed systems projects
• Experience operating systems at telco‑scale or within high‑frequency trading infrastructure

What We Offer

• Opportunity to build voice AI infrastructure that will impact hundreds of millions of Indian users
• Work alongside world‑class AI researchers and engineers on frontier real‑time AI problems
• Access to massive compute infrastructure including next‑gen GPU clusters and edge nodes
• High ownership, fast‑moving culture — no bikeshedding, no monolith thinking
• Competitive compensation and benefits within the company's broader ecosystem