Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m Muzammil Arshad, a Bachelor of Science student in Computing Science at the University of Alberta, with a strong focus on building scalable data pipelines and reliable software systems. I’ve worked on production-oriented analytics and caching strategies, translating large-scale time-series workloads from R/MapR into Python with Polars and PySpark on S3—improving runtime while keeping the system maintainable. I also enjoy building end-to-end prototypes, from a semantic PDF search engine with vector indexing and in-document highlighting to a browser-based full-duplex voice agent. Along the way, I care about correctness, observability, and testability—using practices like partition-aware caching, quality checks, concurrency controls, and automated integration/testing to ship dependable solutions.…I’m Muzammil Arshad, a Bachelor of Science student in Computing Science at the University of Alberta, with a strong focus on building scalable data pipelines and reliable software systems. I’ve worked on production-oriented analytics and caching strategies, translating large-scale time-series workloads from R/MapR into Python with Polars and PySpark on S3—improving runtime while keeping the system maintainable. I also enjoy building end-to-end prototypes, from a semantic PDF search engine with vector indexing and in-document highlighting to a browser-based full-duplex voice agent. Along the way, I care about correctness, observability, and testability—using practices like partition-aware caching, quality checks, concurrency controls, and automated integration/testing to ship dependable solutions.

Muzammil Arshad

Software Engineer, Data Engineer, Voice Actor, +2





I’m Muzammil Arshad, a Bachelor of Science student in Computing Science at the University of Alberta, with a strong focus on building scalable data pipelines and reliable software systems. I’ve worked on production-oriented analytics and caching strategies, translating large-scale time-series workloads from R/MapR into Python with Polars and PySpark on S3—improving runtime while keeping the system maintainable. I also enjoy building end-to-end prototypes, from a semantic PDF search engine with vector indexing and in-document highlighting to a browser-based full-duplex voice agent. Along the way, I care about correctness, observability, and testability—using practices like partition-aware caching, quality checks, concurrency controls, and automated integration/testing to ship dependable solutions.…I’m Muzammil Arshad, a Bachelor of Science student in Computing Science at the University of Alberta, with a strong focus on building scalable data pipelines and reliable software systems. I’ve worked on production-oriented analytics and caching strategies, translating large-scale time-series workloads from R/MapR into Python with Polars and PySpark on S3—improving runtime while keeping the system maintainable. I also enjoy building end-to-end prototypes, from a semantic PDF search engine with vector indexing and in-document highlighting to a browser-based full-duplex voice agent. Along the way, I care about correctness, observability, and testability—using practices like partition-aware caching, quality checks, concurrency controls, and automated integration/testing to ship dependable solutions.

Available to hire

I’m Muzammil Arshad, a Bachelor of Science student in Computing Science at the University of Alberta, with a strong focus on building scalable data pipelines and reliable software systems. I’ve worked on production-oriented analytics and caching strategies, translating large-scale time-series workloads from R/MapR into Python with Polars and PySpark on S3—improving runtime while keeping the system maintainable.

I also enjoy building end-to-end prototypes, from a semantic PDF search engine with vector indexing and in-document highlighting to a browser-based full-duplex voice agent. Along the way, I care about correctness, observability, and testability—using practices like partition-aware caching, quality checks, concurrency controls, and automated integration/testing to ship dependable solutions.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Beginner

Language

Javanese

Intermediate

Aragonese

Advanced

Work Experience

Software Engineer Intern at Ericsson

September 1, 2025 - April 30, 2026

Migrated a statistical time-series analysis pipeline from R/MapR to Python (Polars, PySpark) on S3, using sliding-window anomaly detection on KPI data. Reduced runtime by ~40% while improving scalability and maintainability. Architected a partition-aware L2 caching system integrated with Spark query orchestration, re-querying only missing/changed data and merging results transparently. Added cache correctness guarantees via hash-based invalidation, transient error sentinels, and concurrency control to prevent stale reads and redundant distributed jobs. Built time-consistent ingestion and robust filtering/transformation/quality checks for customer-facing reports, including time-range filters and retry logic for HTTPS API calls. Automated execution using Jenkins and improved reliability with unit and integration testing.

Contingent Worker - Software Engineer Intern at Meta

June 1, 2024 - December 31, 2024

Extended a Python-based silicon validation framework to provision lab and emulation environments and orchestrate automated end-to-end silicon testing and benchmarking workflows. Introduced multi-threaded execution to increase parallel test throughput by 2× across lab and emulation systems. Refactored the execution flow to improve scalability, reliability, and fault isolation in large-scale hardware validation. Identified edge-case failures and performance bottlenecks prior to fabrication, reducing downstream silicon bring-up risk.