Krishna Kumar Singh — DevOps Engineer

I am a Cloud, DevOps, and AI Infrastructure Architect with 16+ years of experience building large-scale, reliable, and automated platforms across FinTech, eCommerce, and SaaS. My expertise spans AWS, Kubernetes (EKS), Terraform, Ansible, CI/CD, Observability, SRE, and most recently LLM/AI Infrastructure and GPU systems.At Paytm, I lead the design and operations of enterprise-grade LLM inference platforms, deploying and optimizing large models such as LLaMA, DeepSeek, and 70B+ parameter LLMs. I specialize in vLLM, Triton, TensorRT-LLM, sglang, and OpenAI-compatible APIs built using LiteLLM. My work includes deep GPU-level optimization (A10/H100/H200), quantization, KV-cache tuning, tensor parallelism, batching strategies, and engineering ultra-low latency (<10ms) multi-model serving pipelines.I architect and manage multi-cluster EKS environments, automated AMI lifecycles (Packer + Terraform), GitOps deployments via ArgoCD/Helm, blue/green rollouts, and zero-downtime upgrades. I also build unified Observability stacks using Prometheus, Grafana, Alertmanager, ELK, and MCP-based AI observability for automated anomaly detection and intelligent RCA.Before Paytm, I scaled infrastructure and automation for Info Edge (Naukri, 99acres, Jeevansathi), implemented CI/CD and hybrid cloud environments for Dineout, and drove AWS modernization for Elara (Housing, PropTiger, Makaan). Earlier roles across Petronia and Infinite Technologies strengthened my foundations in Linux, networking, and high-availability operations.I enjoy building self-healing, cost-efficient, and AI-ready cloud platforms, leading engineering teams, and enabling businesses to adopt next-generation AI and DevOps capabilities with confidence.

Stackforce AI infers this person is a Cloud and DevOps Architect specializing in FinTech and AI Infrastructure.

Location: Greater Delhi, Delhi, India

Experience: 16 yrs 10 mos

Skills

Llm Infrastructure
Devops
Sre
Observability
Infrastructure Automation
Infrastructure Management

Career Highlights

16+ years in Cloud and DevOps architecture
Expertise in AI and LLM infrastructure
Proven track record in cost optimization and automation

Work Experience

Paytm

Paytm — Senior DevOps Manager (AI & Cloud Platform) (4 yrs 4 mos)

Info Edge India Limited. (Naukri.Com Group)

Info Edge India Ltd. — Technical Operations Architect (3 yrs 8 mos)

Times Internet

Dineout (Times Internet) — DevOps Engineer (Dec 2015 – Mar 2018) (2 yrs 3 mos)

Elara Group (Housing, PropTiger & Makaan)

Elara Group (Housing, PropTiger & Makaan) – Member of Technical Staff -DevOps | Dec 2013 – Oct 2015 (1 yr 10 mos)

Petronia Technologies

Petronia Technologies — Technical Team Lead (Oct 2009 – Dec 2013) (4 yrs 2 mos)

Infinite Technologies NCR

Infinite Technologies — Senior Linux System Administrator (Mar 2009 – Sep 2009) (6 mos)

Education

Bachelor of Arts at Dr. Ram Manohar Lohia Avadh University (RMLAU), Faizabad (Ayodhya)

Krishna Kumar Singh

DevOps Engineer

Greater Delhi, Delhi, India16 yrs 10 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

16+ years in Cloud and DevOps architecture
Expertise in AI and LLM infrastructure
Proven track record in cost optimization and automation

Stackforce AI infers this person is a Cloud and DevOps Architect specializing in FinTech and AI Infrastructure.

Contact

Skills

Core Skills

Llm InfrastructureDevopsSreObservabilityInfrastructure AutomationInfrastructure Management

Other Skills

AWX ProjectAlertmanagerAmazon Web Services (AWS)AnsibleArgoCDDocker ProductsEKSELKElastic Stack (ELK)GrafanaGraylogH100/H200/A10 GPUsHelmHelm (Software)Jenkins

About

Experience

Paytm

Paytm — Senior DevOps Manager (AI & Cloud Platform)

Nov 2021 – Present · 4 yrs 4 mos · Noida · On-site

⚡ AI / LLM / GPU Infrastructure
Improved inference success rate by 35%, reduced GPU idle time by 40%.
Designed Paytm’s end-to-end AI & LLM infrastructure, enabling production deployment of large FinTech-focused models.
Built high-throughput inference platforms using vLLM, Triton, TensorRT-LLM, sglang, achieving sub-10ms latency.
Deployed & optimized LLaMA, DeepSeek, 70B+ LLMs with quantization, KV-cache tuning, tensor parallelism & graph optimizations.
Engineered GPU clusters on A10/H100/H200, improving throughput, memory efficiency & token/sec by 40–70%.
Implemented OpenAI-compatible API gateways using LiteLLM for unified model access.
Built MCP-based AI Observability for automated anomaly detection & intelligent RCA.
Achieved ~28% GPU & cloud savings via batching, scheduling & autoscaling optimizations.
Automated AI platform provisioning using Terraform, Packer, Ansible, including GPU node lifecycle.
Tech Stack: vLLM, Triton, TensorRT-LLM, sglang, H100/H200/A10 GPUs, EKS, Terraform, Helm, ArgoCD, Prometheus, Grafana, ELK, Packer, LiteLLM, MCP
☸ Kubernetes & Cloud Automation
Architected and managed multi-cluster EKS with GitOps, Helm, ArgoCD & self-healing deployments.
Automated patching, AMI lifecycle & OS hardening with Packer + Terraform.
Built IaC for EKS, EC2, VPC, RDS, ALB, IAM & S3.
☁ AWS Architecture & FinOps
Delivered 28% cloud cost reduction through GPU right-sizing, autoscaling & compute optimization.
Designed secure VPC, routing & multi-AZ failover.
📊 SRE & Observability
Built unified monitoring using Prometheus, Grafana, ELK, Alertmanager.
Implemented GPU-level metrics for SLO compliance.
👥 Leadership
Led & mentored 10+ Cloud/DevOps engineers.

vLLMTritonTensorRT-LLMsglangH100/H200/A10 GPUsEKS+11

Info edge india limited. (naukri.com group)

Info Edge India Ltd. — Technical Operations Architect

Mar 2018 – Nov 2021 · 3 yrs 8 mos · Noida · On-site

⚙ Hybrid Cloud & Infrastructure
Led production operations across On-prem, OpenStack, Xen & AWS for Naukri, Jeevansathi, 99acres & Shiksha.
☸ IaC & Automation
Automated provisioning of 300+ servers/day using Terraform + Ansible.
Auto-generated CI/CD pipelines tied to infrastructure creation.
🔗 Networking & Load Balancing
Implemented HA microservices routing via HAProxy, LVS, Nginx & Consul.
📊 Observability
Built monitoring using ELK, Grafana, Zabbix.

NginxOpenStackAWX ProjectDocker ProductsHelm (Software)Prometheus.io+12

Times internet

Dineout (Times Internet) — DevOps Engineer (Dec 2015 – Mar 2018)

Dec 2015 – Mar 2018 · 2 yrs 3 mos · Noida · On-site

Managed hybrid infrastructure across AWS and on-prem, supporting mission-critical restaurant discovery and booking platforms.
Modernized application deployments through containerization and microservices using Docker.
Designed and maintained CI/CD pipelines in Jenkins, automating 200+ jobs spanning builds, deployments, continuous testing, backups, and infra diagnostics.
Automated provisioning and configuration using Ansible to ensure reproducible environments.
Implemented HA and load-balanced architectures using ELB, HAProxy, and Nginx to ensure platform stability during peak events.
Tuned performance and improved resilience across services.
Built centralized and scalable logging using Graylog2, enabling quick RCA and multi-environment troubleshooting.
Integrated New Relic for comprehensive APM visibility and anomaly detection.
Implemented SonarQube for automated code-quality checks across engineering teams.
Worked closely with Development and QA teams to streamline release cycles, reduce deployment times, and improve delivery reliability.
Supported production operations, scaling initiatives, and day-to-day infra stability.

AWX ProjectDocker ProductsAnsibleTeam ManagementDevOpsElastic Stack (ELK)+6

Elara group (housing, proptiger & makaan)

Elara Group (Housing, PropTiger & Makaan) – Member of Technical Staff -DevOps | Dec 2013 – Oct 2015

Dec 2013 – Oct 2015 · 1 yr 10 mos · Noida · On-site

Automated end-to-end build and release pipelines using Jenkins, improving deployment speed and reliability.
Managed AWS environments (EC2, ELB, ASG, RDS, VPC, S3) for high-traffic real-estate platforms.
Achieved 30% AWS cost reduction through automation and right-sizing.
Maintained Tomcat, Apache, Nginx, Solr, and backend stacks with 24×7 availability and zero-downtime rollouts.
Built monitoring using Graylog2, S3, Nagios, and New Relic; handled RCA, incidents, and performance tuning.
Supported both on-prem and AWS hybrid infra, ensuring scalable, highly available environments.
Coordinated with Dev, QA, and Infra teams for architecture improvements, stable releases, and performance optimization.

Docker ProductsAnsibleTeam ManagementDevOpsElastic Stack (ELK)Amazon Web Services (AWS)+2

Petronia technologies

Petronia Technologies — Technical Team Lead (Oct 2009 – Dec 2013)

Oct 2009 – Dec 2013 · 4 yrs 2 mos · Noida · On-site

Clients & Duration:
Pine Labs — System Admin (2009–2010)
Vriti / Bagittoday — System Admin (2010–2011)
Clickable / Syncapse — DevOps & AWS (2011–2013)
Guavus — Data Center Support (2013)
Ixigo / Indiahomes — DevOps (AWS) (2013)
Key Contributions:
Provided 24×7 production support across multiple enterprise clients.
Managed Linux servers, networking, and hybrid cloud environments.
Automated deployments and system maintenance with scripting.
Built monitoring, backup systems, and handled RCA.
Coordinated client delivery, vendors, documentation, operations.

NginxDocker ProductsTeam ManagementZabbixDevOpsInfrastructure Management

Infinite technologies ncr

Infinite Technologies — Senior Linux System Administrator (Mar 2009 – Sep 2009)

Mar 2009 – Sep 2009 · 6 mos · Greater Delhi Area · On-site

Provided 24×7 technical support for Linux servers and core IT systems.
Managed user, group, and access administration, including quotas and permissions.
Performed data backup and recovery using tar, dump, and filesystem tools.
Installed and maintained local networks and network devices for smooth operations.
Handled Linux service configuration and troubleshooting to ensure system stability.

Nginx