—
—
Available for new projects
.png)
Arunabho Som
↳
I help businesses turn
data into clear decisions.
Featured Projects✦
Explore a curated selection of
Explore a curated selection of
my recent data science projects
Benefit✦
Work with me, and you'll
Work with me, and you'll
see these benefits
Data‑Driven Insights
I turn raw data into clear, actionable decisions — using statistical analysis, modeling, and visualization tailored to your specific goals.
Collaborative Approach
I partner closely with stakeholders at every stage — from scoping the problem to delivering production‑ready models that fit your workflow.
Proven Results
Measurable outcomes — 86% sentiment classification accuracy, 83% CTR prediction accuracy, and analytics platforms supporting 18,000+ profiles.
Machine Learning Models
Production‑ready models in scikit‑learn, TensorFlow, and PyTorch — from feature engineering to deployment.
Data Analysis
Exploratory analysis, customer segmentation, and statistical insights using Python, SQL, and Pandas.
Dashboards
Interactive Tableau, Power BI, and QuickSight dashboards turning metrics into stakeholder‑ready stories.
Predictive Modeling
Forecasting and classification models — vehicle valuation, churn, CTR, and pricing — built for business outcomes.
NLP & Sentiment Analysis
Text classification, topic modeling, and sentiment systems — from data collection through LLM integrations with LangChain.
Big Data Pipelines
PySpark and Databricks pipelines on AWS and Azure — scaling from millions of rows to production workloads.
How It Works✦
A process built for
A process built for
measurable results
01
Discovery
Understanding your goals, data landscape, and what success looks like for the business.
02
Analysis
Exploratory data analysis, cleaning, and statistical review to surface the signal in your data.
03
Modeling
Building, tuning, and validating predictive models — classical ML through deep learning where it pays off.
04
Delivery
Shipping dashboards, models, and documentation your team can use, extend, and trust over time.
Experience✦
My journey in the world
My journey in the world
of data science.
2025 — Present
Toronto, Canada
Data Science Consultant — Beam Data
- Built an automated competitor‑monitoring pipeline using n8n workflows, Python (Playwright, Selenium, Apify, BrightData) for large‑scale scraping, and PostgreSQL — embedded content into a RAG architecture (Chroma, FAISS) with scheduled LLM summarization to surface pricing changes, product launches, and competitive positioning shifts.
- Delivered Power BI dashboards for Awegoo, an Amazon aggregator covering 200+ brands — tracking inventory health, listing performance, MAP compliance, sales velocity, and SKU‑level profitability.
- Engineered a Vehicle Valuation Model in Python, scikit‑learn, and TensorFlow for a vintage car auction house, trained on 50‑year scraped auction data to drive inventory pricing and sale forecasting.
- Built predictive price models and Tableau dashboards for a luxury watch broker, scraping reference, marketplace, and auction data and processing on Azure to guide retail‑portfolio acquisition.
- Developed AI agents using OpenAI/Anthropic APIs and LangChain to automate research, drafting, scheduling, and engagement workflows for social media operations.
2022 — 2025
Toronto, Canada
Data Analyst, Taxpayer Services — Canada Revenue Agency
- Delivered SQL‑based reports, dashboards, KPI trackers, and data‑validation workflows for a federal contact‑centre network handling 25–32M calls and 16–17M unique callers annually, supporting a program administering $379B in tax revenue and $46B in benefits.
- Applied EDA, trend analysis, predictive analytics, and pattern recognition to taxpayer, benefits, and service‑delivery data — informing workload planning, coaching priorities, and continuous improvement of program‑level accuracy (92%) and professionalism (96%) scores.
- Handled sensitive taxpayer data under Income Tax Act and Excise Tax Act confidentiality provisions, applying strict need‑to‑know controls and audit traceability across every reporting output.
2020 — 2022
Toronto, Canada
Data Analyst — Epiq Global
- Conducted financial and operational analysis in Python and SQL on claims, transactions, and seasonal trends; built customer segmentation models using clustering to drive engagement and retention strategy.
- Built interactive Tableau dashboards for financial KPIs, customer behaviour, churn, and sentiment — delivering real‑time insights that improved service quality and decision cycles.
2019 — 2020
St. John's, Canada
Data Analyst, Insurance — GreenShield
- Managed 18,000+ member profiles in SQL and Excel; applied financial modelling to claim premium structures, claim patterns, and risk profiles to support underwriting and policy decisions.
- Conducted profitability analysis and reserve estimations in SQL, evaluating claim frequency and severity for strategic financial planning and risk assessment.
Inspiration✦
You can't connect the dots looking forward; you can only connect them looking backwards. So you have to trust that the dots will somehow connect in your future.
— Steve Jobs
It can be hard to trust in the process when you can't see the bigger picture. But you never know what might be around the corner, so you have to keep moving forward. And one day, you may recognize that some of the hardest things you had to go through were also the best things that ever happened to you.