Noel John
Data engineering · Machine learning · LLM apps · Analytics
Vancouver, BC
2nd worldwide Ocean Hackathon 20251st place PowerTrust GenAI Hackathon
Experience
- Jan 2026 – Present
Data Analyst (Co-op), Global Operations · Geotab, Vancouver
- Disproved the prevailing theory for a revenue projection overshoot by reconciling the gap against live data; the review owner then made my monthly back-test a standing process.
- Explained a large, recurring gap between two revenue systems by reconciling it piece by piece, showing nearly all of it was a definition difference and pinning the real defects to specific lines of code.
- Designed and built the data-quality layer for the revenue pipelines, so problems surface before they reach the monthly review and dashboards.
- Built a Cloud Run service that diagnoses pipeline failures automatically and posts a root-cause diagnosis to the team.
- Traced a false alarm in an anomaly check to a backfill, proved the data was not corrupted, and showed the check itself was fragile.
- Aug 2021 – Aug 2024
Data Analyst / Business Analyst · Media.net, Mumbai
- Automated the team's ETL workflows in Airflow and Alteryx, cutting manual workload by 40% and the monthly reporting cycle from six hours to under one.
- Cut the typical investigation into a production data issue from eight hours to two.
- Built Python and SQL pipelines over 10M+ records a day across BigQuery and Redshift.
- Built customer lifetime value models covering five mobile apps with 5M+ monthly active users.
- Cut new-analyst onboarding time by 30% through the data-model and validation documentation I kept for the team.
- Built Tableau dashboards and helped move the data warehouse from Redshift to BigQuery.
- Jan 2025 – Dec 2025
Global Recreation Ambassador · Northeastern University, Vancouver
- Organised 20+ recreation events for 150+ international students.
- Jan 2025 – Apr 2025
Teaching Assistant · Northeastern University, Vancouver
- Ran Python exercises, SQL tutorials and Tableau labs for 20 graduate students.
- May 2023 – May 2024
Instructor · Try Catch Classes, Mumbai
- Trained working professionals in Alteryx, SQL and Tableau as a then-certified Alteryx instructor; 8 in 10 moved into analytics roles within three months.
- Dec 2017 – Mar 2018
Volunteer English tutor · Teach India (British Council and Times of India), Mumbai
- Taught spoken English and job readiness to 30+ adult learners; 5+ found jobs afterwards.
Projects
- Jun – Jul 2026Underwater 3D reconstruction benchmark
A benchmark that checks the shape of underwater 3D reconstructions against the real thing, instead of only checking how nice the pictures look.
My part: The Blender scene pipeline and the reconstruction runs on Northeastern's H200 cluster
Research210 generic vs water-aware 3DGS runs over 3 seeds. Water-aware 3DGS helped only on high-relief scenes in clear to moderately murky water.
- Apr 2026Research assistant for solar investment
A research assistant for solar investment in 10 emerging markets that only answers from official sources, and says so when it doesn't know.
My part: Data collection, agents, retrieval and the app
1st place
- Mar 2026Deal handoff tool for an international law firm
A prototype for Dentons that helps lawyers in different countries hand cross-border deals to each other without losing context. My part was building out the system.
Team project
- Nov 2025Fall-protection safety checker for WorkSafeBC
A team hackathon build that checks work-site photos and video for fall-protection violations and maps each one to the BC safety regulation it breaks. My part was the video pipeline.
Team project
- Oct – Dec 2025Ocean current classifier from radar maps
Our Ocean Hackathon project: a model that looks at an hourly radar map of currents in the Strait of Georgia and guesses what's pushing the water around. Tides, wind, or the Fraser River.
My part: The computer vision model and the data pipeline
2nd worldwide76.2% test accuracy on 1,226 held-out maps.
Hackathons
- Apr 2026PowerTrust GenAI Hackathon
1st place
- Mar 2026
- Oct – Dec 2025Ocean Hackathon 2025
1st in Canada, then 2nd worldwide at the Grand Finale in Brest
- Nov 2025InnoFest 2025
Built the video path for a WorkSafeBC fall-protection checker
- Dec 2024Visualization Hackathon, Northeastern
Led a team of 3 to 2nd of 12 teams
About
I build the data systems teams rely on to make decisions: the pipelines that move the data, the reporting on top of them, and the checks that make sure the numbers can be trusted.
Over the past few years that has meant processing millions of records a day, turning slow manual reporting into automated workflows, and finding out why revenue numbers don't add up, then fixing the logic behind them and adding checks that catch the next one. I also build ML models and LLM tools, two of which took top places at hackathons, including 2nd in the world at the Ocean Hackathon.
I also lead and teach. I've coached working professionals into analytics roles, run graduate labs at Northeastern, led a hackathon team, and organised events for international students. I like owning a problem from the first question through to a fix that lasts, and bringing people along with me.
I've always wondered why people do what they do. Why more people prefer one kind of food. Why we open the fridge every hour even though we know there's nothing new in it. It took me a while to realise those are just data points about how people think, and that's more or less how I ended up in this field.
I studied electronics engineering in Mumbai. The subject I did best in was neural networks and fuzzy logic, and one of my undergraduate projects averaged a neural network and a random forest to predict what people would buy next from their invoice history.
Then I joined Media.net. Before I started I thought I knew everything, and that I'd aced the interview. I was wrong. I spent weekends learning SQL, Python and Tableau properly, and a VP there showed me a completely different way to look at a problem. The thing I took away: nothing is really difficult. It just needs more effort, more knowledge, or a different way in.
In 2024 I moved to Vancouver because I didn't want to spend my whole life in one city when there's so much out there to learn.
I started swimming at five and was scared of the deep end, until an instructor threw me in while I was crying and asking for a float. I got over the fear that day and got good at it. Since then I've tried to give most things at least one go before saying no.
I chose this field because it combines the two things I love most, solving problems and understanding how people think. It is something I want to keep doing for a long time.
Education
- Sep 2024 – Apr 2027
MSc, Data Analytics Engineering · Northeastern University, Vancouver · GPA 3.85
- Aug 2017 – Jun 2021
B.E., Electronics Engineering · University of Mumbai
Skills
- Languages
- Python, SQL, R
- Data engineering
- BigQuery, Redshift, PostgreSQL, MySQL, SQLite, Airflow (Cloud Composer), Alteryx, data modelling, data-quality testing
- Cloud and infra
- Google Cloud (Cloud Run, IAM), AWS (S3, Lambda), Azure (Functions, Blob Storage, Video Indexer), Docker, Kubernetes, Linux, SLURM on H200 GPUs
- ML and statistics
- scikit-learn, PyTorch, TensorFlow, XGBoost, H2O AutoML, MLflow, SHAP, regression, clustering, time series forecasting, A/B and hypothesis testing, cohort and lifetime value analysis
- LLMs and NLP
- RAG, multi-agent pipelines, LangChain, LangGraph, FAISS, ChromaDB, spaCy, sentence-transformers, prompt guardrails
- Computer vision and 3D
- EfficientNet, 3D Gaussian Splatting, NeRF, Blender scripting
- BI and analytics
- Tableau, Power BI, Looker, Superset, Excel, Mixpanel, Amplitude, Google Analytics
- Apps and tools
- Streamlit, FastAPI, Flask, Plotly, Git, GitLab CI, GitHub Actions, pytest, Jira, Confluence