Krishna Durai
Software Engineer, Machine Learning at Meta
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolKrishna Durai is a machine learning-focused software engineer with 10+ years of experience building production ML systems, MLOps, and data platforms across startups and large tech firms. After earning his degree from NIT Nagpur, he helped architect AI platforms at SigTuple, led Kubeflow’s on-premise working group (presenting at KubeCon NA 2019), and drove award-winning ML features at Cisco’s Contact Center AI. At Meta he provides technical leadership combining ML research and engineering to ship scalable product solutions, and his open-source contributions include improving Kubeflow manifests and Dex authentication integrations to simplify secure, observable deployments. He blends hands-on DevOps, infrastructure automation, and model engineering expertise, and often operates at the intersection of observability, authentication, and ML workload orchestration—skills honed across diverse geographies from Bangalore to San Francisco.
10 years of coding experience
11 years of employment as a software developer
High School Science, High School Science at Rajhans VidyalayaBachelor's Degree Computer Science and Engineering, Bachelor's Degree Computer Science and Engineering at Visvesvaraya National Institute of TechnologyHigh School, High School at Gokuldham High School
Hindi, English, Tamil
Stackoverflow
111reputation
7kreached
0answers
3questions
Github Skills (30)
kubernetes10
authorizations10
docker10
kustomize10
configuration-management10
kubeflow10
authentication10
user-authentication10
environment-variables10
dockers10
cicd10
go10
prometheus10
jsonnet10
authorization10
Programming languages (12)
Github contributions (5)
kubeflow/manifests
Nov 2019 - Aug 2020
A repository for Kubeflow manifests
Role in this project:
DevOps Engineer
DevOps EngineerContributions:8 reviews, 18 commits, 51 PRs in 9 months
Contributions summary:Krishna primarily focused on automating the deployment and configuration of authentication and authorization components within the Kubeflow ecosystem. They configured Istio and Kustomize for managing authentication and authorization on the `ml-pipeline` service. Furthermore, the user worked on adding and integrating key components such as Dex, Keycloak-gatekeeper, and cert-manager with self-signed certificates, adjusting manifests and tests to streamline deployments. This involved restructuring directory structures and parameterizing configuration files.
kubeflowk8skuberneteskustomizemanifests
kubeflow/kubeflow
Jun 2019 - Apr 2020
Machine Learning Toolkit for Kubernetes
Role in this project:MLOps Engineer
MLOps EngineerContributions:9 commits, 17 PRs, 278 comments in 10 months
Contributions summary:Krishna primarily contributed to the configuration and integration of monitoring and metrics within the Kubeflow environment. Their work involved adding Prometheus annotations, port configurations, and service definitions for components like the tf-operator and pytorch-operator. These changes ensure the collection and exposure of metrics, improving the observability and manageability of machine learning jobs within Kubernetes. The user also updated test configurations to align with these changes.
pythondata-sciencenotebookmachine-learningmlops
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.