Skip to content
View Gaurav1112's full-sized avatar

Block or report Gaurav1112

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Gaurav1112/README.md

Kumar Gaurav

Backend engineer — Java/Spring, Go, Kafka, PostgreSQL, Kubernetes. Bengaluru.

I like finding the defect nobody noticed: two code paths that should be equivalent, where one silently skips a step. Then proving it with a failing test before anyone has to take my word for it.

Open-source contributions

Merged work in projects I use every day:

Project Contribution
vitest ⭐ 17k #11102--maxWorkers was never propagated to projects (merged) · #11109 — same defect class for --hookTimeout (open)
spring-kafka ⭐ 2.5k #4594createContainer(String...) skipped the factory configuration its sibling applied; landed as e40a9797 with backports to 4.0.x and 4.1.x · docs fix #4599 (merged)
spring-amqp #3606 — constructor called an overridable setter instead of its final counterpart (GH-3605, merged)
micrometer ⭐ 4.9k #7896 — proxy JMS session results using the method's declared return type (open)

Projects

  • notification-platform — horizontally scalable SMS/email/push platform designed for 50M users. Java 17 · Spring Boot · Kafka · PostgreSQL · Valkey. Eleven Maven modules, 406 tests green without Docker, zero credentials to run. The interesting part is what happens when providers fail, Kafka redelivers, and workers crash mid-send.
  • rewardgate — deterministic auditor for coding-agent benchmarks. Found that 210 of 500 SWE-bench Verified tasks trip at least one defect check, and replicated the result on 2,438 held-out SWE-Gym instances. Every verdict is proven by execution, not opinion.
  • careerfill — enter your career information once; any job-application form is understood, filled safely, and left for you to submit.
  • guru-sishya — interview-preparation platform: structured topic ladders, spaced repetition, Feynman-style self-explanation drills.

Popular repositories Loading

  1. order-management order-management Public

    E-commerce order lifecycle backend — Spring Boot 3, JPA, automated status transitions, full audit history.

    Java 1

  2. ProductSVC ProductSVC Public

    Product data microservice — Spring Boot, MongoDB, LRU caching, Flyway, Swagger, Dockerized one-command setup.

    Java 1

  3. guru-sishya guru-sishya Public

    Interview-preparation platform — structured topic ladders, spaced repetition, Feynman drills.

    Astro 1

  4. rewardgate rewardgate Public

    42% of SWE-bench Verified is defective (210/500), measured deterministically for $0.00 — one checker matches a published figure to within 2 instances. Audits benchmark tasks before they train a mod…

    Python 1

  5. notification-platform notification-platform Public

    Horizontally scalable SMS/email/push notification platform designed for 50M users — Java 17, Spring Boot, Kafka, PostgreSQL, Valkey. 11 Maven modules, 406 tests green without Docker.

    Java 1

  6. nse-trading-lab nse-trading-lab Public

    Stock screening & backtesting platform for NSE/BSE — 11 strategies with look-ahead protection, Zerodha cost model, swing screener, web dashboard.

    Python