Looking for a job

Manavpreet Singh

@manavpreetsingh

Message

Backend Engineer specializing in event-driven, multi-tenant SaaS reliability and high-throughput integrations.

India

Message

What I'm looking for

I’m looking to build and harden reliable, event-driven backend platforms—owning high-throughput pipelines, driving reliability standards, and improving incident response with measurable uptime and performance gains.

I’m a Backend Software Engineer with 4 years of experience architecting reliable, event-driven systems for multi-tenant SaaS products. I combine deep, hands-on ownership of high-throughput pipelines with cross-team technical leadership—defining engineering standards, driving reliability initiatives, and raising the quality bar across integration-heavy backend infrastructure.

I’ve scaled production systems 20x under load, including building Thena’s Slack-native event ingestion backbone and improving throughput from 5K to 100K+ daily events with zero message drops at 500K peak load. I’ve also delivered measurable outcomes like reducing p99 latency from 30 seconds to under 2 seconds, cutting unplanned downtime by 70%, and improving incident response from multi-hour events to under 30 minutes through runbook-driven protocols and distributed tracing.

Experience

Work history, roles, and key accomplishments

Current

Software Engineer II

Current

Thena

May 2025 - Present (1 year 1 month)

Architected and owned Thena’s Slack-native event ingestion backbone, scaling throughput from 5K to 100K+ daily events (20x) while sustaining zero message drops at 500K peak load. Led a reliability overhaul using Redis NX-key deduplication, DLQ-backed async recovery, idempotent SQS handling, and distributed tracing—improving pipeline availability to 99.9% and reducing mean time to recovery to under

Event Driven Architecture SQS Redis DLQ Based Recovery Reliability Engineering

Full Stack Developer

Peakmind

May 2022 - May 2025 (3 years)

Eliminated API performance bottlenecks, reducing p99 response times from 30 seconds to under 2 seconds at peak (93%) and enabling 195% growth in active users. Built AI-powered conversation flows with OpenRouter, vector search, and RAG pipelines for real-time crisis detection, supporting life-safety intervention for 7 students and cutting unplanned downtime by 70%.

TypeScript NestJS PostgreSQL Redis OpenRouter Vector Search Performance Optimization Incident Reduction