Systems that hold everything together.
I build system for query engines, storage, and data platforms. These quiet layers provide low-level infrastructure for products.
Query Evaluation Engine
I led the design of Liquid’s distributed declarative query evaluation layer, allowing customers to describe what they want without specifying how the system should compute it.
- Hydra: graph-structured in-memory engine for intermediate query states.
- Planner: transforms declarative logic into optimized low-level operations.
- Constraints: enforce semantic correctness and path consistency.
Storage Engine
I co-designed Liquid Storage Layer, a scalable and memory-efficient backend for LinkedIn’s distributed graph query engine.
- Compound indexing: optimized for real-world graph relationships with frequent mutations.
- Branching: supports snapshot isolation for query execution.
- Compaction: improves read efficiency without blocking writers.
- Tiering: balances fast lookups, materialization, and selective access.
Metric View Query Rewrite
I co-designed an intelligent query acceleration layer for AI/BI that rewrites metric view queries to read from pre-aggregated materializations instead of recomputing from raw source data.
- Canonicalization: normalizes query plans for reliable matching.
- Materialization discovery: finds candidate pre-computed views at query time.
- Rewrite strategy: selects the best available snapshot for faster dashboard and analytical workloads.