Worked on enhancing HDFS integration within the ClickHouse/ClickHouse repository by implementing erasure coding support for the pread operation. This involved updating the libhdfs3 submodule to a newer revision and introducing PositionStripeReader.cpp to the build system using CMake. The changes enabled EC-aware reading paths, allowing ClickHouse to perform reliable and efficient data access in HDFS clusters configured with erasure coding. By focusing on C++ development and leveraging expertise in build systems and distributed storage, the work improved read-path reliability and resilience, addressing the needs of data-heavy workloads that depend on robust HDFS erasure coding capabilities for high-throughput environments.