Suraj Kushwah

Work Experience

OnePlay Cloud Gaming - DevOps Engineer

Mumbai, India; July 2023 - Present

• Cloud Infrastructure & Cost Optimization: Architected and meticulously managed production, QA, and development environments across AWS, Linode, and GCP, implementing strategies that significantly optimized cost efficiency without compromising performance or reliability.
• GPU-based Gaming Server Design: Spearheaded the design and optimization of high-performance GPU-based gaming servers, specifically deployed at Yotta Data Center, Mumbai, to deliver a seamless cloud gaming experience.
• Kubernetes Orchestration: Engineered and managed robust Kubernetes clusters, providing scalable and redundant platforms for microservices and critical workloads, enhancing overall system stability.
• CI/CD Implementation: Championed the implementation of comprehensive CI/CD pipelines using Jenkins, automating build, testing, and deployment processes to accelerate release cycles and ensure continuous delivery.
• Cloud Security & Secrets Management: Deployed and configured advanced security solutions including AWS GuardDuty, WAF, and HashiCorp Vault, fortifying cloud security posture and establishing a secure framework for secrets management.
• Observability Stack Development: Built and maintained a sophisticated observability stack utilizing Grafana, Prometheus, Kibana, and Zabbix, integrating custom alerting mechanisms for proactive system health monitoring.
• Database Cluster Management: Proficiently managed and optimized diverse database clusters, including Percona MySQL, MongoDB, Redis, InfluxDB, and Elasticsearch, ensuring data integrity and high-speed access.
• Automation & Infrastructure as Code: Automated server provisioning, scaling, and maintenance routines leveraging Terraform, Ansible, and custom Bash/Python scripts, significantly reducing operational overhead.
• Incident Management & Resilience: Led on-call incident response, implementing proactive monitoring and automated failover strategies that demonstrably minimized downtime and enhanced system resilience.

Bipolar Factory - DevOps & Backend Engineer

Bangalore, India; November 2021 - April 2023

Falcon - AI Powered Surveillance

• Real-time CCTV Inference Pipelines: Developed high-efficiency real-time CCTV inference pipelines, leveraging DeepStream & TensorRT, resulting in substantial improvements in video processing speed for AI-driven surveillance.
• AI-based Smart Surveillance API: Constructed a resilient REST API for AI-powered smart surveillance, seamlessly integrating automated anomaly detection and proactive security alerts.
• Distributed Processing Optimization: Optimized distributed processing pipelines, ensuring robust handling of high-resolution video feeds under demanding operational conditions.

Goa Election Project

• Real-time Election Monitoring: Managed and monitored over 200+ RTSP CCTV streams, providing real-time oversight for election processes, demonstrating capability in high-stakes environments.
• AI-based Vote Counting System: Developed an innovative sound-based vote counting system, leveraging AI to accurately detect voting machine beeps, enhancing election integrity.
• Automated Server Deployment: Automated server deployment and scaling mechanisms, crucial for effectively monitoring high-traffic election dashboards during critical periods.

StreamQuicker - Low-Latency Video Streaming

• WebRTC to RTMP Streaming Service: Developed a low-latency WebRTC to RTMP streaming service, integrating FFmpeg, RTMP, and ossrs/srs to ensure efficient video transmission.
• Auto-scalable Distributed Architecture: Designed and implemented an auto-scalable, distributed streaming architecture, significantly improving streaming latency and overall reliability.
• AI-driven Load Distribution: Utilized advanced AI models to predict and intelligently distribute streaming loads across servers, optimizing resource utilization and system responsiveness.

AiViS - Industrial AI & Automation

• AI-powered Defect Detection: Developed a sophisticated AI-powered defect detection system, enhancing manufacturing quality control processes through automated inspection.
• PLC Integration for Automation: Integrated PLC controllers to facilitate real-time industrial automation within factory conveyor systems, demonstrating expertise in OT environments.
• Data Pipeline for Defect Analysis: Constructed a robust data pipeline for comprehensive analysis, alerting, and reporting of defects in production lines, ensuring proactive quality assurance.

SafePlant - ML-Based Factory Workers Safety System

• ML-based Safety System Development: Led the conceptualization and development of a plant safety system, leveraging machine learning and computer vision to autonomously detect safety hazards.
• TensorFlow Model Training: Trained advanced machine learning models using TensorFlow to meticulously analyze plant images and videos for the identification of potential risks.
• Flask Backend with WebSocket: Developed a Flask-based backend web service, integrated with WebSocket technology, enabling real-time communication for immediate safety alerts and monitoring.