AI/ML Engineer - AI Observability (ML Ops Engineer)

Vconstruct; Pune District, Nagpur District; On-site; INR 2,400,000 - 4,200,000; Full time

Job summary

Vconstruct is seeking an AI Observability Engineer in Pune to build scalable, cloud-native observability frameworks for our AI systems. You will standardize logging, metrics and model monitoring across teams and lead automation-first DevOps/MLOps initiatives to enable reproducible environments.

You will design DataOps pipelines, deploy containerized ML workloads, extend CI/CD for ML workflows, and establish governance for bias, explainability and auditability.

Qualifications

• Bachelor's degree in CS/DS or related field.
• 5+ years in DevOps, MLOps, Data/Software/SRE roles.
• Proficiency in Python with ML frameworks like TensorFlow, PyTorch or Scikit-learn.

Responsibilities

• Standardize observability across AI/ML teams including logs, metrics, tracing and model monitoring.
• Lead automation-first DevOps and MLOps practices, enabling IaC and reproducible environments.
• Design intelligent DataOps pipelines with automated data quality monitoring and anomaly detection.
• Deploy, maintain and monitor containerized ML workloads.
• Extend CI/CD for automated infra changes and ML workflows.
• Govern AI systems with bias detection, explainability and auditability.
• Automate RBAC, role and permission management to reduce manual work.
• Develop automated tests for model performance, regression and bias validation.
• Monitor KPIs: accuracy, precision, recall, latency, calibration.

Skills

DevOps & MLOps

Data Engineering

Software Engineering

Site Reliability Engineering

Education

Bachelor's degree in CS/DS/Info Systems or related field

Tools

Azure

Python

TensorFlow

PyTorch

Scikit-learn

Job description

Position Overview

The AI Observability Engineer will be instrumental in implementation of scalable, cloud-native solutions to meet the growing needs of our Data & Development team. The successful candidate will demonstrate the ability to abstract complexity and create reusable, scalable patterns that accelerate development. The AI Observability Engineer will build and maintain a robust framework to ensure the reliability and maintainability of DPR Construction's complex AI systems.

Responsibilities

• Standardize observability practices across AI/ML and other development teams including logging, metrics, tracing, and model performance monitoring, ingesting data from multiple platforms
• Lead hands-on implementation of automation-first DevOps and MLOps practices, enabling infrastructure-as-code and consistent, repeatable environment provisioning
• Design and manage intelligent DataOps pipelines with automated data quality monitoring and anomaly detection
• Deploy, maintain and monitor containerized ML workloads
• Extend existing CI/CD pipelines to support automated infrastructure changes and ML workflows
• Implement AI-driven data validation, schema and concept drift detection and metadata management.
• Establish governance frameworks for AI systems, including bias detection, explainability, and auditability
• Extend existing Azure RBAC strategy by automating role and permission management to reduce manual intervention
• Develop automated test suites for model performance, regression, edge cases and bias validation
• Monitor model KPIs (accuracy, precision, recall, latency, calibration)
• Ensure reproducibility of experiments and production models
• Act as a technical point of contact for DevOps and MLOps practices, developing reusable patterns, documentation, and proof-of-concepts to drive adoption

Qualifications

• Bachelors degree in computer science, Data Science, Information Systems, or a related field
• 5+ years of experience in DevOps, MLOps, Data Engineering, Software Engineering or Site Reliability Engineering
• Strong understanding of cloud infrastructure and experience working with at least one major cloud provider, preferably Azure
• Proficiency in at least one objected-oriented programming language, preferably python with hands-on experience in ml frameworks like TensorFlow, PyTorch or Scikit-learn