Vijai Jayavel
Infra Dev Specialist @Cognizant
Signup Ā· Get unlimited contacts
WORK HISTORY
Infra Dev Specialist @Cognizant
Coimbatore, IN
Architected production LLM solutions using Vertex AI, Gemini Enterprise & LiteLLM on GCP. Built multi-agent & subagent systems for enterprise automation using Google Agent frameworks. Enabled enterprise chatbot platforms integrating LLMs with internal APIs & knowledge bases. Implemented AI Governance frameworks for responsible and compliant LLM deployments. Built LLMOps observability using Arize Phoenix ā monitoring token consumption, cost/query, latency, hallucinations & model drift in production. Designed Elastic Dashboards for LLM cost tracking ā token usage, API spend & per-team chargeback reporting. Established LLMOps CI/CD pipelines for model deployment & prompt versioning. Led multi-cloud IaC on AWS & GCP using Terraform with S3/GCS backends; built reusable modules for consistent provisioning. Managed Jenkins lifecycle ā CloudBees upgrades via Helm, migrated to open-source to cut costs & vendor lock-in. Built shared Jenkins libraries & GitHub Actions workflows for scalable, reusable CI/CD pipelines. Deployed self-hosted GitHub Actions runners as Kubernetes pods for ephemeral, scalable builds. Integrated SonarQube & Veracode security scans into CI/CD pipelines. Managed AWS & GCP deployments with health checks, load balancers & auto-scaling; automated compute shutdowns with Python to reduce costs. Created Ansible playbooks for legacy server deployments; configured multi-language build tools (Maven, Gradle, npm, PyPI). Built SSL expiry & PagerDuty alert workflows; automated Jenkins & PagerDuty user audits for AD compliance.
EDUCATION
Bishop Francis Matriculation School
SSLC, Matriculation
Sri Ramakrishna Mission Vidyalaya Polytechnic College
Diploma, Information Technology
Sri Ramakrishna Engineering College
Bachelor of Technology, Information Technology
SKILLS
ABOUT VIJAI JAYAVEL
Senior DevOps | LLMOps | SRE Engineer with 10+ years of experience building and operating cloud-native infrastructure ā now at the forefront of AI/LLM platform engineering.š¤ Currently designing and deploying production-grade LLM solutions ā working with Vertex AI, Gemini Enterprise, LiteLLM, and Arize Phoenix for end-to-end LLM observability. Building multi-agent and subagent architectures, enabling enterprise chatbots, and driving AI Governance frameworks. Built LLMOps observability platforms using Arize Phoenix and Elastic Dashboards ā tracking token consumption, cost per query, latency, hallucination rates, and model drift in real-time production environments. Deep expertise in GCP, Terraform, Kubernetes, GitHub Actions, Jenkins, Docker, and Ansible ā bridging the gap between traditional DevOps reliability and modern AI/ML operations (LLMOps). Passionate about AI Governance, responsible AI deployment, and ensuring LLM systems are observable, secure, and production-ready. Always exploring: Agentic AI, MCP (Model Context Protocol), and next-gen CI/CD patterns for LLM pipelines. Reach out to me at vā¢ā¢ā¢ā¢ā¢ā¢ā¢ā¢@outlook.com /+91 -
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.