TrueFan - DevOps Engineer - Cloud Infrastructure (2-4 yrs)

Key Responsibilities :

Model Training/Deployment Pipelines and Monitoring :

- Design, implement, and maintain scalable and automated pipelines for deploying deep neural network models.

- Monitor and manage Production models, ensuring high availability, low latency, and smooth performance.

- Automate workflows for data preprocessing (face alignment, feature extraction, audio analysis), model retraining, and video generation.

- Implement Logging, Tracking, and Monitoring Systems to ensure data integrity and visibility into the model lifecycle.

Infrastructure Management :

- Build and manage cloud-based infrastructure (AWS, GCP, or Azure) for efficient model training, deployment, and data storage.

- Collaborate with DevOps to manage containerization (Docker, Kubernetes) and ensure robust CI/CD pipelines using github and jenkins for model delivery.

- Monitor resource for GPU/ CPU-intensive tasks like video processing, model inference, and training using Prometheus , Grafana, alert manager, ELK stack.

Collaboration :

- Work closely with ML engineers to integrate models into production pipelines.

- Provide tools and frameworks for rapid experimentation and model versioning.

Required Skills :

- Basic Python

- Strong experience with cloud platforms (AWS, GCP, Azure) and cloud-based machine learning services.

- Expert knowledge of containerization technologies (Docker, Kubernetes) and infrastructure-as-code
(Terraform, CloudFormation)

- Have understanding of Deployment of both synchronous and asynchronous API using Flask, Django, Celery, Redis, RabbitMQ , Kafka

- Deployed and Scaled AI/ML in Production.

- Familiarity with deep learning frameworks (TensorFlow, PyTorch).

- Familiarity with video processing tools like FFMPEG and Dlib for handling dynamic frame data.

- Basic understanding of ML models

Preferred Qualifications :

- Experience in image and video-based deep learning tasks.

- Familiarity with media streaming and video processing pipelines for real-time generation.

- Experience with real-time inference and deploying models in latency-sensitive environments.

- Strong problem-solving skills with a focus on optimising machine learning model infrastructure for scalability and performance.