Distributed Drip Rate-Limiter for Event Streams
Built a distributed drip rate-limiting worker and reusable scheduler package to throttle GCP Pub/Sub streams into deterministic time windows using Cloud Tasks and overflow topics.
Read case studyOpen to lead, staff, and principal engineering roles
Manas Yadav
Lead Engineer · Distributed systems & cloud platforms
I architect high-throughput distributed backends and event-driven systems that scale 10x–100x — transforming complex infrastructure into cost-efficient, high-reliability platforms that directly accelerate business growth.
- 7+ years building scalable backends
- 10x–100x system scaling & cost optimization
- Architecture from startups to global platforms
About
Engineering scale with business impact
Over my years in backend and platform engineering across HealthTech, FinTech, and SaaS, I’ve contributed in diverse domains and built systems both from scratch and at massive throughput. My focus has beed deep technical execution with clear business outcomes — scaling backend architectures from 10x to 100x, optimizing cloud spend to bring down system costs, and turning fragile services into bulletproof operations.
As a senior developer and technical lead, I look at systems proactively with a keen eye for architectural bottlenecks before they turn into outages. I’ve guided engineering teams through the intense challenges of high-growth startups, placing high value on mentoring developers, elevating engineering standards, and building resilient platforms that drive long-term business value.
Velocity gets attention, but reliability earns trust. The best architecture scales both product ambition and operational confidence.
Background
Experience & expertise
- 2024 – Present
Lead Engineer HighLevel India Pvt Ltd
Lead a 4-person platform team driving platform reliability, security, and developer infrastructure. Drove company-wide Go adoption and rewrote our core IAM AuthN/AuthZ service in Go, improving performance by 70%. Built and scaled a low-latency secrets manager from 0 to 200M+ calls/day. Also created platform-wide shared packages that unlocked deep observability across teams, and currently building an agent-driven IDP for automated performance analysis.
- Go
- NestJs
- MongoDB
- GCP
- GCP PubSub
- Firestore
- Elastic Search
- AI Agents
- 2023 – 2024
Software Development Engineer III CoinDCX
Spearheaded Project Mercury in Go and gRPC, architecting event-driven Portfolio, PnL, and multi-country wallet systems. Scaled real-time WebSocket streaming across Okto and CeFi with incremental updates, cutting DTO costs by 60%. Led a zero-downtime cross-region Kafka migration that reduced broker latency from 250ms to <10ms for 15M+ users.
- Go
- gRPC
- Kafka
- WebSockets
- Redis
- PostgreSQL
- Event-Driven Architecture
- 2022 – 2023
Software Development Engineer II CoinDCX
Re-architected core wallet infrastructure to introduce on-demand wallet generation and led optimization across an interdependent 22-billion-row dataset via vacuum tuning and archival—slashing DB size by 80%, IO costs by 60%, and query latency by 40%. Concurrently engineered the Okto Web3 token pricing engine with Neo4j and Redis Cluster, halving API latency and cutting DTO transfer costs by 60%.
- Ruby on Rails
- PostgreSQL
- Neo4j
- Redis Cluster
- AWS DMS
- Kafka
- 2021 – 2022
Software Development Engineer I CoinDCX
Built the internal state store using event sourcing on Kafka to capture high-throughput trading data with reliable state recovery. Migrated market data services out of the monolith into modular microservices with Express and TypeScript, cutting API latency by 40% and streamlining downstream trade ingestion.
- TypeScript
- Node.js
- Kafka
- PostgreSQL
- BullMQ
- Event Sourcing
- 2020 – 2021
Full Stack Developer Somnoware Healthcare Systems Pvt Ltd
Jun 2020 – Jun 2021 · Built core product features and platform capabilities for sleep-disorder diagnosis, healthcare data compliance, and clinical testing workflows.
- 2019 – 2020
Software Developer CronJ IT Technologies
Jun 2019 – Jun 2020 · Developed backend microservices and client-facing features while helping modernize monolithic codebases into scalable, service-oriented architecture.
Core stack
Case studies
Work that shows impact
Production systems across fintech, healthcare, and platform engineering — with measurable reliability and scale outcomes.
lockIt: High-Throughput Low-Latency Encrypted Storage
Designed a centralized storage mechanism to hold encrypted user-sensitive data. Handled 300M+ calls daily with sub-millisecond latency, optimized for 80% read-heavy traffic patterns.
Read case study
Zero-Downtime Cross-Region Kafka Migration
Migrated our core Kafka cluster from the US to India with zero downtime, cutting publish and consume latency from 250ms to under 10ms for live trading systems.
Read case study
Stitch
Energy-guided developer dashboard built with Go, Wails, and React to manage GitHub repositories, PR reviews, and build pipelines via local GitHub CLI auth.
View repository
GoKafkaToolkit
Toolkit for Kafka producers and consumers across local, cloud, and Confluent environments.
View repository