Helping startups and scale-ups architect, build, and optimize data infrastructure for millions of users. From CDC pipelines to real-time analytics — I make your data work.
About Me
I'm Manikaran Kathuria — a Data Platform Engineer and Consultant with 6+ years of experience building scalable, reliable, and cost-efficient data infrastructure from scratch. Currently working as an SDE-3 across backend and data infra domains.
I specialize in helping startups and scale-ups solve their hardest data challenges: from designing CDC pipelines and real-time analytics stacks to optimizing billion-row databases and cutting infrastructure costs. My work sits at the intersection of systems engineering and data architecture.
M.Tech, IIT Delhi
Computer Science, 2019
6+ Years
Software & Data Infra
India
Working Remotely
Services
Targeted consulting offerings for companies that need to build, fix, or scale their data infrastructure.
Problem
No existing data infrastructure or a fragmented, unreliable stack.
Approach
Design and build an end-to-end data platform with ingestion, processing, storage, and serving layers using proven open-source tools.
Outcome
A reliable, scalable data platform that grows with your product and team.
Problem
Stale data, manual ETL, or broken sync between operational and analytics databases.
Approach
Implement Change Data Capture with Debezium, Kafka, Kinesis, and Flink for real-time data flow with custom offset management.
Outcome
Near-instant data freshness across all systems with zero-downtime syncing.
Problem
Slow dashboards, analytics queries timing out, or no self-serve analytics.
Approach
Deploy fast OLAP engines like StarRocks alongside data lakes (S3 + Iceberg) and dbt for transformation.
Outcome
Sub-second analytics queries and empowered teams making data-driven decisions.
Problem
Production queries running for minutes, slow API responses, or database bottlenecks.
Approach
Deep-dive into query plans, add targeted indexing, implement partitioning and partitionwise aggregation strategies.
Outcome
Up to 1000x improvement in query performance. Measurable, provable results.
Problem
Cloud data bills growing faster than your revenue. Wasteful compute and storage.
Approach
Audit query patterns, optimize storage formats, right-size infrastructure, and implement intelligent caching.
Outcome
Hundreds of dollars saved daily without sacrificing performance or reliability.
Problem
Silent pipeline failures, missing data, or no visibility into system health.
Approach
Build comprehensive monitoring, alerting, and observability across the entire data stack.
Outcome
Proactive issue detection, faster incident resolution, and data you can trust.
Technical Stack
Case Studies
Designed and built an end-to-end data lake architecture on S3 with Apache Iceberg, coupled with a real-time analytics stack using StarRocks. This replaced a fragmented system of ad-hoc queries and slow batch jobs.
Key Results
Identified and optimized critical production queries on billion-row PostgreSQL tables through systematic analysis of query plans, strategic partitioning, composite indexing, and partitionwise aggregation strategies.
Key Results
Built a query intelligence layer that analyzed data access patterns, identified wasteful compute, optimized storage formats, and right-sized infrastructure across the entire data platform.
Key Results
Blog
Technical deep-dives and lessons from the trenches.
A deep-dive into architecting a reliable, cost-efficient data lake from scratch using Apache Iceberg on S3.
Read on MediumCDCLessons learned from implementing CDC pipelines with custom offset and state management in production.
Read on MediumPerformancePractical techniques for partitioning, indexing, and partitionwise aggregation on billion-row tables.
Read on MediumGet in Touch
Have a data challenge? I'd love to hear about it. Whether you need a full platform build-out or a targeted optimization sprint, let's talk.