backend engineer who likes making things fast and not break in production.

building microservices for a utility platform that processes ~25k transactions/day. mostly fighting with databases, optimizing queries, and making APIs go brrr.

just got APIs from 2000ms → 500ms by actually using indexes properly (shocking, i know). also migrated 450k records from ancient SAP system without burning everything down.

// backend stuff
Java 17, Spring Boot, REST APIs
Microservices (the good kind, not the buzzword kind)
JWT, OAuth, all that auth jazz

// databases
MongoDB (with actual indexes this time)
PostgreSQL, Redis
Query optimization is my cardio

// cloud & ops  
AWS: EC2, S3, SQS
Docker (containerize all the things)
CI/CD pipelines that actually work

// testing
JUnit, Mockito, JMeter
load testing until it breaks, then fix it

200+ REST APIs across billing, payments, auth modules
handling 50k+ requests/day with <500ms latency

payment integrations for PhonePe & Cashdesk
because processing money is slightly important (99.5%+ success rate)

zero-downtime migration of 450k records
SAP → MongoDB without anyone noticing

75% faster APIs through:

• compound indexes (they're not just for show)
• Redis caching the right way
• query optimization that actually matters

message queues with AWS SQS
async processing so servers don't die during peak load

10k+ SMS/day through third-party integrations
real-time notifications that actually arrive

PDF generation for bills & receipts
Apache PDFBox doing the heavy lifting

if it's not tested, it doesn't work. wrote tests that caught 200+ bugs before they hit prod.

obsessed with performance metrics. if i can't measure it, i can't optimize it.

production issues get fixed fast. <2hr MTTR because nobody likes being paged at 3am.

code reviews are for learning, not ego battles.

• distributed systems (CAP theorem is wild)
• kubernetes orchestration (yaml hell but worth it)
• system design patterns (trying to not reinvent wheels)
• better observability (logs are just the beginning)

• linkedin: shubham-kumar-tripathi
• email: [email protected]
• location: noida, india (open to remote)

down to chat about:

• why your API is slow (probably missing indexes)
• microservices architecture (and when NOT to use them)
• scaling databases without crying
• production war stories
• that one bug you've been hunting for 3 days

pro tip: if your query takes >100ms, add an index. if it still takes >100ms, add a better index. if it STILL takes >100ms... maybe rethink your schema.

currently caffeinated and debugging