A computer science engineer building data pipelines, ML systems and real-time processing for teams that actually measure what they ship. Currently at CommentSold; writing clean, maintainable code and the occasional essay.
Computer Science Engineer by training, a pipelines person by practice. Currently a Machine Learning & Data Engineer at CommentSold, working on big-data tooling, ML pipelines and real-time processing.
Over the past few years I've deepened my expertise through hands-on projects and online courses — seeking not just knowledge but wisdom. I believe in building meaningful products, not just features: every line of code carries the potential to help or hurt a business, so I commit to clean, maintainable, scalable solutions and regular refactoring to keep technical debt honest.
Outside of code: I'm a tech enthusiast who loves exploring new tools, a keen interest in AI, ML and data science, and a fondness for thoughtful debate — well-chosen discussions spark growth and new perspectives.
Working with big data technologies, ML pipelines, real-time data processing and building data pipelines at scale. Blending discipline with curiosity to ship clean, maintainable systems that earn their keep.
Built and scaled data pipelines and APIs for analytics workloads. Four years of hardening the unglamorous middle of the stack — the part that decides whether the graph goes flat or keeps paging people at 2 a.m.
Actively working through the program — prioritising depth of understanding over a rushed completion. Goal is to master the concepts before sitting the final capstone exam.
Deepened expertise across Machine Learning, Artificial Intelligence and Data Science — the theoretical grounding that makes the pipelines at work less mysterious.
Solid foundations in the core of computer science. Honours track, systems and algorithms focus.
Context-repair layer for production RAG. Catches hallucinations via citation cross-checks, stitches in missing sources.
Debezium → Iceberg with schema-aware merges. Handles 60M row/hour without blinking.
Tiny Rust SSG built around Pandoc. Incremental, typed front-matter, ~2k LOC. Powers this site.
Local-first notes w/ CRDT sync. Markdown, backlinks, full-text, no server when you don’t want one.
Near-real-time feed analytics for a 200M-user consumer app. Spark structured streaming + ClickHouse.
Personal dotfiles. Curated, minimal, boringly-reliable. 200+ stars somehow.
Pulled live from Medium · @codingcerebrum. Click through to read on Medium — the canonical source for everything I publish in long form.