Principal Engineer - Nurix AI

Meraki Labs; Bengaluru; On-site; INR 3,500,000 - 6,500,000; Full time

Job summary

Nurix AI, based in Bengaluru, seeks a Principal Engineer to architect and scale its production AI infrastructure. You will lead distributed systems, ensure low-latency real-time inference for voice and chat, and drive secure, compliant deployments across multi-cloud environments.

You will mentor engineering teams, evaluate new tech (SSMs, Triton, Riva, vLLM), and align research with production readiness. A hands-on leadership role focused on architecture and performance.

Qualifications

• 10–15 years of experience in large-scale systems architecture and at least 5 years in principal architect-level roles.
• Proven expertise in distributed systems, cloud-native architectures, and real-time pipelines.
• Hands-on experience with containerization, orchestration (Kubernetes), and microservices.
• Strong background in scalable ML infrastructure, including model serving, GPU/accelerator utilization, and CI/CD for ML.
• Demonstrated ability to architect systems with low latency (<300ms), high throughput, and enterprise reliability.
• Experience in conversational AI, speech systems, or real-time inference workloads.
• Deep knowledge of MLOps platforms (Kubeflow, MLflow, VertexAI, SageMaker).
• Familiarity with state-of-the-art inference optimization frameworks (Triton, Nvidia Riva, vLLM, SGLang).
• Open-source contributions or patents in distributed systems, infra, or ML tooling.

Responsibilities

• Design and evolve end-to-end infrastructure supporting ASR/TTS, LLM orchestration, Agentic RAG, and self-learning workflows.
• Architect low-latency pipelines for real-time conversational AI with sub-second response times.
• Build multi-cloud, distributed systems (AWS, GCP, Azure) with elastic scaling for spiky workloads.
• Define and enforce SLAs around latency, uptime, and throughput for AI services.
• Drive observability, monitoring, and resilience strategies to handle failures gracefully.
• Optimize GPU/TPU utilization for cost-effective training and inference.
• Partner with InfoSec to embed security-by-design across AI/ML workloads.
• Translate cutting-edge research into production-grade platforms; mentor engineering teams.

Skills

Distributed systems

Cloud-native architectures

Kubernetes

Low-latency design

ML infrastructure

Real-time pipelines

Security & compliance

Leadership & mentorship

MLOps platforms

Inference optimization

Tools

Kubeflow

MLflow

VertexAI

SageMaker

Triton

NVIDIA Riva

vLLM

SGLang

Job description

Job Description:

About Nurix AI

At Nurix AI, we are pioneering the Autopilot Enterprise. Our conversational AI agents handle workflows, drive outcomes, and deliver measurable impact for businesses. Born from the belief that enterprises need a new playbook, we build autonomous, multilingual agents capable of complex reasoning, contextual understanding, and end-to-end workflow ownership. Backed by $27.5M in funding from Accel, General Catalyst, and Meraki Labs, and led by Mukesh Bansal, we are India's first scaled enterprise AI company, delivering cutting-edge AI solutions that integrate seamlessly into workflows across industries like Retail, Insurance, Education & Home Services. Join us in shaping the future of enterprise AI — where every interaction is smarter, faster, and human-like.

About The Role

As Principal Engineer at Nurix AI, you will be the cornerstone of our technical infrastructure, enabling our AI agents to scale reliably and securely in production. You will design and oversee distributed systems that deliver low-latency, high-availability voice and chat AI, while meeting enterprise-grade security and compliance requirements. This is a hands‑on leadership role focused on architecture, systems design, and performance engineering — ensuring that Nurix's groundbreaking AI research translates into robust, real‑world deployments.

Key Responsibilities

Systems Architecture & Scalability

• Design and evolve the end-to-end infrastructure supporting ASR/TTS, LLM orchestration, Agentic RAG, and self-learning workflows.


• Architect low-latency pipelines for real-time conversational AI, ensuring sub-second response times across voice and chat.


• Build multi-cloud, distributed systems (AWS, GCP, Azure) with elastic scaling to handle spiky workloads.

Reliability & Performance Engineering

• Define and enforce SLAs around latency, uptime, and throughput for AI services.


• Drive observability, monitoring, and resilience strategies to handle failures gracefully.


• Optimize GPU/TPU utilization for cost-effective training and inference.

Security & Compliance

• Partner with InfoSec to embed security-by-design across all AI/ML workloads.


• Implement controls to protect sensitive enterprise data while meeting global compliance standards (SOC2, ISO 27001, GDPR, DPDP).

Collaboration & Leadership

• Work closely with the Head of AI to translate cutting-edge research into production-grade platforms.


• Provide technical mentorship to engineering teams, ensuring best practices in distributed systems and infra design.


• Evaluate and adopt emerging technologies (e.g., SSMs, inference optimizers like Triton, Riva, vLLM) to stay ahead of the curve.

Required Qualifications & Skills

• 10 - 15 years of experience in large-scale systems architecture, with at least 5 years in principal architect-level roles.


• Proven expertise in distributed systems, cloud-native architectures, and real-time pipelines.


• Hands-on experience with containerization, orchestration (Kubernetes), and microservices.


• Strong background in scalable ML infrastructure, including model serving, GPU/accelerator utilization, and CI/CD for ML.


• Demonstrated ability to architect systems with low latency (<300ms), high throughput, and enterprise reliability.


• Experience in conversational AI, speech systems, or real-time inference workloads.


• Deep knowledge of MLOps platforms (Kubeflow, MLflow, VertexAI, SageMaker).


• Familiarity with state-of-the-art inference optimization frameworks (e.g., Triton, Nvidia Riva, vLLM, SGLang).


• Open-source contributions or patents in distributed systems, infra, or ML tooling.