Yash Sachdeva Email & Phone Number

San Francisco Bay Area

San Francisco, California, Us

- Designed multiple Release Management strategies to cater to the different product offerings. Have designed and developed a CI/CD and release framework for Mono-Repo and Multi-repo products, including handling multi-tenant services.- Helped design a multi-tenant Infrastructure tech stack to support a Enterprise AI SaaS Product with highly scaleable compute - Designed scalable infrastructure to support thousands of Data Science training and inference tasks over 500+ compute nodes on AWS + Kubeflow- Currently designing and building a tenant management platform to tackle the entire lifecycle of a customer, from onboarding, setting up the infra, product deployment, to offboarding.- Designed and developed a robust Infrastructure and Application Monitoring Solution using Datadog

San Francisco, California, Us

San Francisco, California, Us

- Extensive experience with Kubernetes on both Microsoft Azure and AWS. This includes Microsoft Azure AKS, AWS EKS as well as bare metal Kubernetes. Managed production workloads of over 500+ Data Science pipelines on Kubernetes clusters in Microsoft Azure- Designed a central alerting, logging and monitoring system using Elasticsearch, Kibana and Logstash. Working to migrate to a new system comprising Cloudwatch, Prometheus and Grafana- Designed the Continuous Integration and Delivery Process to be used across the organization using Jenkins. Currently evaluating GitOps tools such as Argo CD and Flux- Worked on standardizing the Data Science model training workflows, at scale, using frameworks such as Kubeflow and MLFlow. Developed a model deployment service for Data Science models- Helped build a novel MLOps framework, which aims at standardizing the Data Science development and deployment processes including both model training as well as inference. Contributed towards the development of the CLI for this framework- Currently working to design a scalable infrastructure backend using AWS EMR (Elastic Map Reduce) with Spot Instances for a historical Data Processing Framework. Worked with Data Engineers to standardize the Big Data Cluster (Hadoop, Spark, Yarn and Ambari)- With the help of Terraform, Ansible and Python, created an auto infrastructure provisioning tool leveraging Infrastructure as Code principles. This is now the standard for Infrastructure setup for all projects in the organization- Worked with project leads to create a DevOps process that helped reduce the application migration from development to production time from 3 weeks to 4 days

Cambridge, Ma, Us

Worked as an AWS and Linux professional in the Cloud Operations department at Pegasystems. Specialized in AWS, Linux Operating System, maintaining, troubleshooting web and application servers and managed customer expectations.

Munich, De

Building management systems division, system controller design for air conditioning and fire safety systems.

Bachelor Of Engineering - Be, Electronics And Communications Engineering

High School, Higher Education/Higher Education Administration