Job Title: ML Ops Engineer – Speech AI & Infrastructure (5–8 Years)

Location: Bangalore (India)

Full-Time | Immediate Joiners Preferred

Role Summary:

We are looking for an experienced ML Ops Engineer (5–8 years) with a strong background in operationalizing AI/ML systems, particularly in speech AI agents integrated into telephony environments. This role bridges AI infrastructure, voice bot deployment, and real-time streaming operations, enabling AI models to function reliably in production use cases such as customer collections, voice-based load bookings, and conversational AI in logistics and telecom.

Key Responsibilities:

• Build, deploy, and maintain real-time speech AI agents for telephony (inbound/outbound) using CPaaS platforms and audio pipelines.
• Integrate ASR (Automatic Speech Recognition), TTS (Text-to-Speech), and LLMs with voice bot workflows using WebSocket or RTP streams.
• Work closely with backend engineers and AI scientists to deploy scalable ML models into production across Kubernetes, Docker, and cloud-native environments (AWS/GCP).
• Set up observability, logging, and monitoring of speech pipelines (latency, dropouts, stream integrity) using Prometheus, Grafana, ELK, etc.
• Automate the ML lifecycle and CI/CD for AI models using tools like MLflow, Airflow, or Kubeflow.
• Manage streaming latency optimization across ASR, LLM, and TTS chains in low-latency applications like telephony bots.
• Ensure fault-tolerant, secure, and compliant deployment of voice-based systems using industry-standard DevOps practices.

Required Skills:

• 5–8 years of experience in ML Ops, AI platform engineering, or DevOps with hands-on ML deployment.
• Experience with ASR/TTS model integration (e.g., Whisper, Amazon Polly, Google Speech API).
• Hands-on experience deploying AI in telephony environments, with working knowledge of Asterisk, FreeSWITCH, or similar systems.
• Solid command over WebSocket, SIP, RTP, or similar streaming protocols for voice integration.
• Proficient with cloud services (AWS/GCP/Azure), containers (Docker), orchestration (Kubernetes), and infrastructure as code (Terraform).
• Experience with end-to-end CI/CD pipelines, GitOps, and model monitoring (accuracy, drift, performance).
• Strong programming in Python and scripting to support infrastructure and integration tasks.

Nice to Have:

• Exposure to CPaaS platforms (e.g., Twilio, Vonage, Kaleyra, Exotel) for voice bot integration.
• Familiarity with conversational AI/NLU platforms and agentic AI frameworks.
• Prior experience supporting AI agents in collections, logistics/freight, or telecom workflows.

What We Offer:

• Work on next-gen AI voice agents deployed in high-impact customer environments.
• Influence platform architecture and deliver cutting-edge infrastructure for real-time ML.
• High-ownership, engineering-led culture with strong product alignment and technical mentorship.