Data & BI Engineer · 3+ years

Surbhi Bhor

I build data systems that turn raw data into scalable pipelines, executive dashboards and real decisions.

3+ years across the full data lifecycle, from serverless automation and cloud ETL to analytics dashboards that inform strategy at scale.

Open to work
Based in India · open to remote
IST · overlaps EU and US-East mornings

surbhi.bhor.in@gmail.com · linkedin.com/in/surbhi-bhor · github.com/surbhi-bhor · surbhi-bhor.runs-on.dev

Email meResume ↓LinkedIn ↗GitHub ↗

In numbers

incremental revenue drivenWhole Foods Market: $5.9M
shorter ad-hoc reporting cyclesWhole Foods Market: 71%
of data discrepancies eliminatedBoston Consulting Group: 75%
shorter pipeline runtimeCalimontex: 30%

Selected work

Whole Foods MarketAmazon Grocery

Business Intelligence Engineer · Dec 2024 – Aug 2025 · Austin, TX

The problem

Retail decisions at Whole Foods Market waited on slow reporting: supplier renewals, return-counter staffing, a $130M dry-grocery seasonal business.

What I did

Business Intelligence Engineer for three Whole Foods Market programmes: Supplier Marketing, Parcel Returns and Seasonal. I owned their data from AWS ingestion through Redshift marts to the QuickSight dashboards the business ran on.

Results

$5.9M in supplier entitlements backed by campaign reporting
71% shorter ad-hoc reporting cycles, with dashboards refreshing in under a minute
About $20K a year saved in AWS compute through Spark tuning

Spark SQL
Redshift
AWS Glue
QuickSight
Lambda
S3
Python

Read the full case →

Calimontex Technologies

Data Engineer · Jan – Aug 2026 · Austin, TX

The problem

Finance, merchandising and the stores each brought their own number to the same meeting. Core retail metrics meant something different in every report.

What I did

Data Engineer on the warehouse behind those reports. I owned its rebuild end to end: the data model, the pipelines and the lineage behind every core metric.

Results

30% shorter pipeline runtime with incremental processing
One definition of net sales, margin and stockout rate across every report
Self-serve metric lookup, so analysts no longer ask engineering where a number came from

SQL
Python
Cursor
Star schema
Data lineage
Incremental ETL

Read the full case →

Boston Consulting Group

Data Engineer Intern · Jun – Aug 2023 · New York, NY

The problem

A client running gas stations and convenience stores had records arriving from many systems that stopped agreeing long before production.

What I did

Data Engineer Intern on the client’s Azure platform. I built the layer that takes raw transactions to trusted, promotion-ready data.

Results

75% of data discrepancies eliminated with record-level PySpark validation
Full and incremental CDC loads in Synapse across the Medallion layers
A 15+ table domain modelled, and a shared design tool adopted by the team

Azure Synapse
ADLS Gen2
PySpark
Data Factory
Delta Lake
SCD
CDC

Read the full case →

Content Intelligence Pipeline

Personal project · Self-hosted data platform

The problem

Useful signals scattered across four APIs, with no warehouse to hold them and no tests to vouch for them.

What I did

A self-hosted ELT platform for personal movie, TV and book history. Data from four APIs (Simkl, Hardcover, TMDB, OpenLibrary) lands in PostgreSQL, orchestrated by Dagster and modelled in dbt with incremental facts, idempotent loads and pipeline guardrails. Flask and Metabase serve it, and a local Llama 3.2 turns plain-English questions into SQL and picks ten movies and five books every Friday.

Results

Four APIs (Simkl, Hardcover, TMDB, OpenLibrary) land in one PostgreSQL warehouse
Orchestrated in Dagster and modelled in dbt, with incremental facts and idempotent loads
A local Llama 3.2 answers plain-English questions in SQL, so no question leaves the machine

Dagster
dbt
PostgreSQL
Python
Flask
Metabase
Ollama · Llama 3.2
LangChain
Docker

See the code on GitHub ↗

LearnLoop

Personal project · Agent skill for Claude and Notion

The problem

I learn most technical topics by chatting with AI. Within a week most of it was gone: the places I got stuck disappeared, and I re-asked the same questions weeks later without noticing.

What I did

An Agent Skill that connects Claude to Notion. Say "log this" at the end of a learning chat and it turns the conversation into a fully formatted study note: where I got stuck, a diagram, flashcards, interview questions for the target role, a practice drill and a review date. It also remembers past doubts, so when the same confusion returns under a new topic, it asks me to recall the answer before explaining again.

Results

Note-taking time per topic cut from 60–90 minutes to about 6 (90–93% less)
Nine manual steps, from formatting to flashcards to review reminders, became one command
Caught the same doubt across 3 different topics and asked me to recall it first

Agent Skills
Notion MCP
Python
Claude
Claude Code
Codex

See the code on GitHub ↗

Bright Mind Enrichment

Data Engineer · pro bono · Aug – Dec 2024 · Austin, TX

The problem

A nonprofit's chapters tracked volunteer hours, donations and outreach in a patchwork of spreadsheets and intake forms. Monthly reporting took about twenty hours.

What I did

Centralised the data into an OLAP structure and built dashboards connecting Street Care app, volunteer and donation KPIs, so chapter leads pull their own numbers instead of waiting for someone to compile them.

Results

Monthly reporting cut from about 20 hours to under 2
Dashboards for 10+ chapter leads, who now pull their own numbers
12,600+ people aided, and every one of them counted

SQL
OLAP
Tableau
ETL

Capgemini

Software Development Engineer in Test · Sep 2020 – Jun 2022 · Mumbai, India

The problem

A quick-service restaurant’s ordering system kept adding endpoints and markets, and its testing had not kept pace.

What I did

Software Development Engineer in Test on that platform. I built automated API testing and the monitoring around it, and was picked for the team that grew API automation from one person into a full group.

Results

70% code coverage and 200+ automated test cases across 7 microservices
40% faster release cycles with testing built into every pipeline
Issues resolved in under 6 hours instead of 2+ days, with mean time to resolution halved

Node.js
AWS Lambda
Step Functions
API Gateway
CloudWatch
Power BI
Azure DevOps

Read the full case →

Kind words

From managers, teammates and the business teams I built for.

Surbhi quickly made an impact by showcasing the ability to independently manage her own work and her own stakeholders. If and when the opportunity arises, I would 100% hire Surbhi again.

Nick HayesAnalytics Leadership, Whole Foods Marketmanaged Surbhi directly

What impressed me most about Surbhi was her ability to transform ambiguous business requirements into functional, impactful solutions with minimal guidance. Surbhi consistently delivered high-quality work ahead of schedule, keeping our entire team on track with project deliverables.

Matt LarsonSenior BI Engineer, Whole Foods Marketworked alongside Surbhi

Surbhi has always been helpful and responsive when we run into tech issues or discrepancies. She is open to implementing updates and has been patient. I appreciate her partnership with our team!

Anna OlsonProgram Manager, Supplier Marketing, Whole Foods Marketbusiness stakeholder

You have been a great partner. I appreciate your quick responses and taking time to explain when issues have popped up.

Andi DowdaGlobal Marketing Director, Whole Foods Marketbusiness stakeholder

Surbhi excelled at organizing large datasets, improving data quality, and using advanced data techniques. Her analytical skills are top-notch, and she’s great at simplifying complex data for the team.

Yashwanth Reddy ChegireddyCloud & Data Platform Lead, BCG Xmentor at BCG

Her strategic thinking was evident when she evaluated various ERD migration tools, leading to a well-informed decision that enhanced real-time collaboration and saved considerable time. She led data modeling efforts effectively, creating intuitive ER diagrams and data flow diagrams for multiple pods, making intricate concepts more accessible.

Toby LeungSenior AI Engineer, BCG Xworked alongside Surbhi

She rarely ever came to me with a problem without a solution in hand. So to say it was my privilege to have worked with her, to have led her, would be underselling it.

Sidhant MishraSenior Software Engineer, Capgeminimanaged Surbhi directly

Skills

Languages & frameworks: SQL · Python · PySpark · Spark SQL · Pandas · JavaScript · Node.js
AWS: Redshift · Glue · Lambda · S3 · QuickSight · Step Functions · API Gateway · CloudWatch · DynamoDB
Azure: Synapse Analytics · ADLS Gen2 · Blob Storage · Data Factory · Delta Lake · Azure DevOps
Data & BI: PostgreSQL · MySQL · dbt · Airflow · Dagster · Tableau · Power BI · Metabase · Excel · Git · Postman
AI-assisted tooling: Cursor · Claude · GitHub Copilot · Amazon Qfor script refactoring, data transformation and pipeline debugging

Education

University of Illinois Urbana-ChampaignMaster of Science, Information Management · GPA 3.9/4.0Aug 2022 – May 2024
University of MumbaiB.E., Electronics EngineeringAug 2016 – Oct 2020

Contact

The next case is yours.

I'm currently open to Data Engineering, Business Intelligence Engineering and Analytics Engineering roles where I can own the full data lifecycle, from pipeline architecture to the dashboards people actually use.

Email meResume ↓LinkedIn ↗GitHub ↗