Bengaluru, Karnataka, India
I build the systems other engineers ship on: CI/CD release gates, cloud device farms, event-driven validation services, and observability pipelines on AWS. At Amazon Music, I work across the full engineering stack: I've shipped a pre-merge validation service (SNS → Lambda → CodeBuild with PrivateLink artifact resolution), a production CD release gate orchestrated via Step Functions, infrastructure-as-code in AWS CDK for pipelines and backend services, and an autonomous self-healing AI agent for engineering workflows. My work is less about writing tests and more about building platforms where release decisions are trustworthy and developers move faster. What I work with: • Event-driven backend services on AWS (Lambda, Step Functions, EventBridge, ECS, SNS) • Infrastructure-as-code (AWS CDK), CI/CD platforms, release gating • Cloud device infrastructure: 15+ devices, 100+ parallel executions • Observability pipelines (Elasticsearch, CloudWatch): debug in minutes, not hours • Practical GenAI: autonomous agents, MCP servers, LLM-assisted triage (~50% faster sev-2 debugging) Outside work, I've mentored 550+ engineers (https://topmate.io/sahil_kapoor) through DSA, system design, and interviews for top-tier roles. Teaching the craft keeps me sharp at it. Tech: Python · TypeScript · Java · AWS · CDK · Docker · Kubernetes · Redis · GraphQL · Elasticsearch · distributed systems · LLM agents If you want a pragmatic review of a flaky pipeline, CI architecture, or AI-agent workflow please DM me.
1. Designed and shipped an event-driven pre-merge validation service: code-review events fan out via SNS → Lambda → CodeBuild, provisioning a Spring Boot service + Redis, resolving real build artifacts over VPC PrivateLink, and posting pass/fail verdicts back to the review, moving regression detection before merge. 2. Built a production CD release gate orchestrated via Step Functions with a deterministic exit-code contract, blocking unstable builds from deployment. 3. Authored infrastructure-as-code (AWS CDK) for production pipelines and backend services, ECS, Redis, API Gateway, CloudFront. 4. Built and operated cloud device infrastructure: 15+ devices across 3 servers, automated registration/de-registration, 100+ parallel executions across OS/device variants. 5. Diagnosed and fixed a ~90-minute build-time regression in the Android app pipeline, built a baseline-profile pipeline (publisher Lambda + CDK) to accelerate app startup. 6. Integrated 700+ backend API tests into dev/prod approval workflows as automated quality gates, dramatically improving early feedback. 7. Built an org-wide observability pipeline (Elasticsearch → Lambda → dashboards, SES/JIRA/Slack outputs) exposing system health across Amazon Music teams. 8. Built a VegaOS (next-gen Fire TV) performance framework capturing UPL & telemetry across 7 KPIs (TP90/TM90), from POC to handed-off production framework. 9. Led a GenAI triage initiative (logs + context → root-cause summaries), cutting developer debug time ~50% on sev-2 incidents; shipped a self-healing autonomous AI agent and an MCP server adopted by the team.
1. Enhanced the automation test infrastructure, improving CI pipeline speed and reliability, small changes, big impact. 2. Took ownership of a robust email automation framework using Google APIs, built from scratch, and it worked like a charm. 3. Optimized the core automation framework to improve performance and scalability across projects, faster, cleaner, better. 4. Jumped in to resolve critical testing issues, ensured smooth releases and no surprises in prod. 5. Set quality benchmarks for test cases and frameworks, consistency matters, and we made it stick. 6. Led initiatives with strong ownership and a solution-first mindset, got things done, end-to-end. 7. Proactively handled incident-based automation requests, delivered quick, reliable solutions that unblocked teams. 8. Contributed to functional correctness by designing critical test cases, especially for P0s and high-risk features. 9. Improved flaky test reliability and helped stabilize key jobs like qa-browser-test and qa-multi-deploy. 10. Simplified complex flows like property setup, file uploads, and window handling with reusable helpers. 11. Collaborated closely with cross-functional teams, always open to feedback, always learning. 12. Documented processes and helped shape best practices, to make life easier for the next person.
1. I initiated my tenure as a member of the Automation team, where I spearheaded the development of an Automation framework from the ground up. Employing various optimization techniques, I ensured the framework's robustness and stability. 2. Pioneered the seamless integration of the comprehensive framework with Twilio's OTP mechanism, facilitating real-time authentication for users during runtime. This innovation enhanced the security and reliability of our automation processes. 3. Played a pivotal role in the DevOps sphere, responsible for the installation, configuration, and integration of automation scripts within the realm of continuous integration tools such as Jenkins. This pivotal effort led to the nightly execution of test suites, ensuring consistent validation of our systems. 4. Orchestrated the end-to-end automation of internal applications including Jira, Confluence, and Bitbucket, core tools utilized across various projects within Publicis Sapient. This endeavour significantly streamlined processes, fostering efficiency and collaboration across the organization.