Abhishek
Kothari — Cloud Solutions Architect & Principal SRE

All projects →

1. Feb 2024 – Present ● Current

   Virtasant

   Principal SRE (Consultant)⌄

   Client: confidential · planet-scale on-prem environments plus smaller ones on AWS.

   Featured projects

   • Fleet upgrade automation — automated upgrades for a fleet of 2000+ machines, with proper draining of hosts and the necessary guardrails built into the automation
   • AI incident response — automated incident root cause analysis and incident response
   • Just-in-time SSH — fleet-wide rollout of just-in-time SSH access to strengthen the security landscape
   • SLO observability — improved the observability landscape for better SLO tracking of services

   AI & automation

   • Build AI-powered RAG pipelines to automate documentation search
   • Create a Slack bot that automatically investigates incidents using document search, metric plugins and log APIs
   • Leverage AI-assisted coding to build interactive TUIs and automation tooling at a fast pace
   • Build automation to reduce SRE toil as and where possible

   Platform & infrastructure

   • Manage observability instrumentation for large-scale self-hosted GitHub and Artifactory
   • Automate infrastructure provisioning and upgrades using Pulumi and Terraform
   • Manage planet-scale environments running on-premise and other smaller ones on AWS
   • Manage infrastructure as code using Puppet, Ansible, Terraform and Terragrunt
   • Instrument dynamic GitHub Actions infrastructure on AWS for Mac-based instances
   • Build GitHub apps & pre-receive hooks to enforce checks on abusive activities for self-hosted GitHub

   Analytics & people

   • Build dashboards tracking growth and usage metrics for various tools
   • Automate analytics-based reporting using ClickHouse and GitHub Enterprise systems
   • Build an infrastructure usage-forecasting backend for analysing usage growth
   • Mentor juniors to get up to speed with the products and perform their best; interview candidates around the world for the SRE/DevOps team
   • Learn quickly and adapt to new projects — regardless of whether they involve coding or SRE work

   AI / RAGIncident automationJIT SSHFleet automationSLOsGitHub EnterpriseArtifactoryPulumiTerraformClickHouseClaude Code
2. T 3 Jan 2022 – Feb 2024

   Twilio

   Principal Software Engineer — L5 (Infrastructure R&D)⌄

   Product: Twilio Segment · systems handling billions of transactions.

   Featured: open-source observability

   • Designed, deployed and operated a high-volume (30M+ active series) observability stack built entirely on open source: Grafana, Prometheus, OpenTelemetry Collector, Grafana Loki and Tempo
   • Migrated the observability tech stack from Datadog to open-source tooling

   Scale & reliability

   • Built highly scalable and available systems (10M+ rps – 99.9% availability)
   • Handled the AWS ECS → AWS EKS migration while maintaining five 9's of availability
   • Re-architected systems handling billions of transactions from a cost-optimisation aspect
   • Architected HIPAA-compliant networking solutions for EKS clusters using Cilium Cluster Mesh

   Automation, IaC & people

   • Automated SRE processes such as upgrades, migrations and provisioning
   • Wrote Terraform IaC for the infrastructure solutions above
   • Built and customised open-source tools like K2, kubeapply and others
   • Mentored teams of software engineers through the SRE & DevOps journey to help them grow

   AWS EKSCiliumGrafanaPrometheusOpenTelemetry CollectorLokiTempoTerraformHIPAA
3. 15 Jun 2020 – 31 Dec 2021

   Apple

   Software Engineer (Cloud Platform Engineer)⌄

   Remote · product confidential.

   • Developed and maintained applications and utilities using Java Spring, Python, Docker, Kubernetes and Git
   • Architected easily scalable and cost-optimised solutions for cloud platforms
   • Led a team of consultants in development of assigned modules
   • Participated in and planned critical design decisions for scalable data-management platforms
   • Reviewed and improved architectures from security and cost aspects
   • Managed deployment servers for the assigned modules
   • Automated code deployment and installation of supporting tools for virtual machines
   • Managed and resolved issues related to AWS cloud infrastructure and Kubernetes infrastructure
   • Created Helm charts for automating bulk deployment of multiple projects using an automated code pipeline
   • Developed IaC using Pulumi & Terraform libraries

   Java SpringPythonDockerKubernetesHelmPulumiTerraformAWS
4. 5 Jan 2019 – 14 Jun 2020

   RocketML

   Remote Backend Developer (Freelance Consultant)⌄

   Product: RocketML — Elastic Cloud Compute Platform.

   Backend & AWS

   • Developed Java Spring Boot APIs for the Angular 8 frontend
   • Leveraged Hibernate, Spring Security, Spring JPA, Spring Data, Lombok and other libraries for a faster, optimised solution
   • Led project execution and planning for feature releases
   • Set up AWS infrastructure for deployment; used the AWS SDK to programmatically manage elastic cloud compute resources
   • Used CloudWatch, Lambda, ECS and EC2 to optimise resource costs
   • Built dynamic AWS IAM role-assignment logic for secure execution of tasks
   • Integrated AWS Polly, Alexa, S3 and RDS with dynamic Linux-based instances
   • Built Docker images for a variety of machine-learning libraries and Amazon Machine Images using Packer
   • Worked with ML and parallel-computing tools such as AWS ParallelCluster, MLflow, Airflow, JupyterLab and RocketML
   • Built AWS CloudFormation templates for simplified deployment across environments

   Azure, GCP & DevOps

   • Enabled Azure customers to use the application — ARM templates, Azure VM, scale sets, DNS and more for dynamic infrastructure
   • Replicated the setup on Google Cloud Platform: Deployment Manager templates, code deployment pipelines, app services and programmatic infrastructure deployment
   • Diagnosed and fixed a variety of network configuration issues
   • Managed DevOps with AWS CodePipeline & GitHub, internally and for customers; created Postman test suites for automated CI/CD tests

   Spring BootAWSAzureGCPPackerCloudFormationDockerML tooling
5. 22 Jan 2018 – 5 Mar 2019

   Streebo

   Tech Lead⌄

   Products: DXA Chatbot Builder and Streebo Mobile Forms.

   DXA Chatbot Builder — smart transactional multi-channel chatbot

   • Led a team of developers managing the flagship product, Streebo Chatbot Builder
   • Proposed and laid down architecture for enhancements that reduce business effort
   • Developed NodeJS backend services (Joi, Sequelize for MySQL, Bluebird, ExpressJS, Handlebars) and Angular 7 widgets leveraging IBM tools
   • Studied NLP engines: Google Dialogflow, Microsoft Bot Framework, Wit.ai, IBM Watson and AWS Alexa
   • Built a smart multi-lingual chatbot on IBM Watson and Google Cloud services; enabled self-learning with IBM Watson Discovery to reduce training effort
   • Built bots for multiple channels with Dialogflow and Google Assistant; integrated the chatbot into Google Home and Amazon Dot
   • Took part in customer implementations and helped banking clients architect chat-support systems

   Mobile Forms — forms runtime

   • Led a team of interns and technical consultants
   • Developed REST web services (JAX-RS, Java) for form creation and data management; studied IBM Forms Experience Builder architecture
   • Developed an LDAP-based authentication and authorization system and OSGi service hooks, adapters and management code
   • Enhanced functionality for better performance; conducted technical interviews for technical and BA roles

   NodeJSAngular 7IBM WatsonDialogflowJava JAX-RSOSGiLDAPMySQL
6. 2 Jun 2016 – 19 Jan 2018

   Oracle

   Applications Engineer — Oracle HCM Fusion⌄

   Oracle India · three product modules.

   US Payroll

   • Managed seed data and fast formulas used for payroll calculation
   • Maintained and enhanced payroll element creation
   • Developed Linux scripts to automate forward-porting of code

   UI/UX

   • Rectified issues in the existing UI
   • Made the UI mobile responsive using the Oracle JET framework

   Core HR

   • Resolved issues in existing Core HR functionality and enhanced it for better performance
   • Took client calls to understand issues and resolve them faster

   Oracle HCM FusionFast FormulasOracle JETLinux scripting
7. 10 Mar 2014 – 27 May 2016

   TCS

   Systems Engineer⌄

   Tata Consultancy Services.

   • Developed a multi-threaded Java application to manage sensor feedback in a vehicle
   • Created Linux scripts to manage threading and program execution
   • Developed NodeJS web services for analysing and monitoring vehicle positions
   • Developed client-side web application modules on a Spring MVC + Hibernate framework
   • Created backend file-processing modules for government offices using the ZK framework
   • Worked in a dynamic environment with tight deadlines; worked at the client side to monitor optimisation issues during project UAT and pilot

   JavaNodeJSSpring MVCHibernateZKLinux
8. E 1 Jun 2013 – 7 Mar 2014

   Edulark

   Software Developer⌄

   Edulark Services — where the career began.

   • Developed the Spring MVC based web application teacheron.com
   • Created Linux scripts to manage threading and program execution
   • Onboarded new employees onto development using the framework
   • Handled production issues with the website

   JavaSpring MVCLinux scripting

Full experience page →

Full skills index →

AWS Certified Solutions Architect – Associate

AWS Certified Developer – Associate

AWS Certified Cloud Practitioner

Oracle Certified Associate, Java SE 8 Programmer I

B.E. Electronics & Communication
Gujarat Technological University · 2009–2013 · CPI 8.52

Certifications →

Read all testimonials →

Dec 01, 202515 min readDistributed Systems

Microservices vs Monoliths: Making the Right Choice

The choice between a monolithic and a microservices architecture is one of the most critical decisions a software engineering team faces. It impacts everything from development velocity and team structure to operational

Dec 01, 202513 min readDistributed Systems

On-Call Best Practices: Preventing Burnout

The relentless hum of production systems, the constant vigilance required to maintain their health, and the inevitable late-night pages are an inherent part of the modern software engineering landscape. For Site Reliabil

Oct 28, 202512 min readCost-Management

Capacity Planning in the Cloud Era

In the dynamic landscape of cloud computing, managing infrastructure effectively is less about static provisioning and more about intelligent, adaptive resource orchestration. Capacity planning, once a periodic, often te

Oct 27, 202514 min readSite Reliability Engineering

Incident Management: Post-Mortem Culture That Works

The world of complex distributed systems is inherently unpredictable. Despite our best efforts in design, testing, and deployment, incidents are not a question of “if,” but “when.” For Site Reliability Engineers, Softwar

Oct 27, 202514 min readSite Reliability Engineering

Observability vs Monitoring: Understanding the Difference

In the rapidly evolving landscape of distributed systems, microservices, and cloud-native architectures, the terms “observability” and “monitoring” are often used interchangeably, leading to confusion and, more criticall

Oct 27, 202511 min readSite Reliability Engineering

Chaos Engineering in Production: A Practical Guide

We’ve all been there. It’s 3 AM, and the pagers are screaming. A critical service is down, customers are impacted, and the on-call team is scrambling through logs and dashboards, trying to piece together a puzzle in the

All 17 articles →