amit kumar.

Business intelligence · Analytics · Automation

Complex data.
Clear decisions.

I’m Amit. I build the reporting, analytical tools and reliable workflows that help teams understand what matters—and act on it.

Experience at Deutsche Börse Group and Arcadis.
Based in Frankfurt. Open to Data & BI opportunities.

Explore my work Download CV
THE PERSON BEHIND THE DATA01 / AK
Amit Kumar
Frankfurt am Main
ANALYST&BUILDER
10+Derivatives categories
covered by BI reporting
20+ hrsSaved monthly through
reporting automation
929Companies in my
equity-market research

01 / Selected industry work

Useful analysis.
Measurable outcomes.

I work across the full problem: understand the decision, organise the data, and build something people can use.

BUSINESS INTELLIGENCE01
10+derivatives product categories

Eurex Clearing AG · Deutsche Börse Group

A clearer view of market performance.

Self-service Power BI reporting and consistent KPIs for commercial and product teams at Eurex Clearing.

Read the approach

The challenge

Commercial and product teams needed consistent, scalable reporting across clearing members, derivatives products and liquidity trends. Static Excel reports created recurring ad-hoc requests and fragmented performance views.

What I built

Built self-service Power BI dashboards and KPI frameworks for derivatives performance across 10+ product categories. Supported the migration from static reporting into a more scalable BI setup using centralized data models.

Outcome

  • 10+ derivatives product categories covered
  • Consistent KPI definitions across commercial and product teams
  • Reduced dependency on static reporting
  • Stakeholder-ready market intelligence
Power BIDAXDatabricksKPI Frameworks
WORKFLOW AUTOMATION02
20+ hrssaved each month

Deutsche Börse AG

Making recurring reporting repeatable.

Reconciliation and reporting workflows that reduce manual processing in cash market operations.

Read the approach

The challenge

Cash market operations involved recurring manual reconciliation, file processing and reporting workflows. These tasks created manual effort, inconsistent quality checks and avoidable operational risk.

What I built

Implemented and improved automation workflows with Power Automate and AI Builder, supported high-volume reconciliation with Power Query and VBA, and created repeatable controls for recurring operational reporting.

Outcome

  • 20+ hours saved per month
  • 50%+ fewer manual errors
  • Repeatable controls for recurring reporting
  • Faster reconciliation and exception review
Power AutomatePower QueryVBAAI Builder
OPERATIONAL ANALYTICS03
15+HS2 work packages supported

Arcadis

Connecting delivery teams to their data.

Power BI dashboards and Power Apps supporting programme visibility and engineering workflows at Arcadis.

Read the approach

The challenge

Large infrastructure teams needed reliable KPI visibility and resource tracking across multiple work packages in the UK HS2 rail programme.

What I built

Built Power BI dashboards, Power Apps solutions and automated data transformation pipelines for engineering and project management stakeholders.

Outcome

  • 15+ work packages supported
  • Led BIM design automation delivery
  • Automated validation and transformation pipelines
  • Faster approval workflows
Power BIPower AppsPythonVBA
Eurex Clearing AG · Deutsche Börse GroupStatistical Outlier Detection for Institutional Fee Analysis

Large analytical datasets required consistent anomaly detection and clearer visibility into the drivers behind unusual changes.

Developed a Python statistical outlier-detection model using SciPy and Statsmodels to surface unusual patterns and support structured driver analysis.

  • Automated anomaly flagging
  • Repeatable statistical review process
  • Faster prioritisation of unusual patterns
  • More consistent analytical review

Python / Pandas / SciPy / Statsmodels

Deutsche Börse AGComputer Vision Document Verification

Document quality checks required repetitive visual review for key elements and exceptions, creating avoidable manual effort and inconsistency.

Trained and optimized an AI-based computer-vision quality-control tool using Python and OpenCV to review documents and flag exceptions for manual review.

  • 400+ documents reviewed monthly
  • Automated exception flagging
  • Reduced repetitive manual review
  • More consistent quality checks

Python / OpenCV / PyMuPDF

02 / The project lab

FEATURED BUILD Updated October 2026

GridShift.

What if the same operation
ran at a different time?

A historical electricity-cost simulator backed by an API-to-dashboard pipeline. Compare feasible operating schedules, check the cost difference and see whether the underlying data is complete.

  • Equal energy. Different start times.
  • Daily cloud refresh with coverage checks.
  • Failed refreshes preserve validated data.
PythonPostgreSQLdbtPrefectStreamlitGitHub Actions
Launch interactive demo Read the case study

Public demo refreshes historical prices daily through GitHub Actions. Freshness and validation status are visible in the app. The full database pipeline runs locally.

GridShiftInteractive example

Same energy. Better timing.

Move a four-hour run within 06:00–18:00.

Price, EUR/MWh Lowest-cost run
Baseline€50
At 08:00€20
Potential difference€30

Same 400 kWh · Continuous four-hour operation

Synthetic prices for illustration. Wholesale energy only; fees and taxes excluded. This example is not a live market feed.

More experiments in applied analytics

EXPLORE THE WORK ↗
02

Revenue & Customer Retention Analytics

Revenue Atlas

Built a commercial analytics workspace that explains recurring revenue movements and customer retention, with traceable invoices across 600 fictional accounts.

PythonStreamlitPandasPlotly
03

Data Engineering

End-to-End Analytics Pipeline

Built a modular ETL pipeline with ingestion, cleaning, validation, aggregation and multi-format exports.

PythonPandasNumPyOpenPyXL
04

Applied Analytics

Data-Driven Portfolio Analysis

Built a Python analysis that measures annualized return and volatility, studies correlation, and simulates 20,000 portfolios to approximate the efficient frontier.

PythonpandasNumPymatplotlibyfinance

03 / Research & perspective

Curious about
what moves markets.

Does climate deregulation reduce business costs—or introduce a new market risk? My master’s research examines that question across US, European and Asian equities.

The study estimates market-model cumulative abnormal returns and uses panel regression to compare equity return and risk responses across regions, sectors and firm characteristics.

1.0Thesis grade
929Firms analysed
R + PythonImplementation

Four policy events. 929 firms in the event study. A regression sample of 899 firms and 3,589 firm-event observations, with explicit checks for event timing and model sensitivity.

Explore the research case study

MASTER’S THESIS · GRADE 1.0 · MANUSCRIPT FOR ZfU

Explore the research question

Do major US climate policy reversal events generate measurable abnormal returns in global equity markets, and do firm-level characteristics or regional exposure explain differences in market reaction?

Sample: S&P 500, STOXX Europe 600, Nikkei 225, Hang Seng and KOSPI 200.

The manuscript reports heterogeneous reactions at the proposed-rule stage rather than a uniform deregulation benefit. Regional estimates depend on event-date and window choices.

FIELD NOTES / 01 OF 03

Markets & Economics

Does an inverted yield curve really predict recessions?

A look at what a market signal can tell us—and where interpretation needs care.

Read the analysis Illustrative yield-curve sketch

04 / The background

From infrastructure
to financial markets.

My work has taken me from engineering and programme analytics at Arcadis to financial-market reporting at Deutsche Börse Group. The common thread is making complex information useful to the people working with it.

The full story in my CV
BI & reporting

Power BI · DAX · SQL · Databricks · Fabric

Automation & analysis

Python · Power Automate · R · Statistical modelling

Independent project stack

PostgreSQL · dbt · Prefect · Streamlit

Jan 2026 - Present

Eurex Clearing AG

Business Analytics Intern, Market & Liquidity Analytics

Eschborn, Germany

Jul 2025 - Dec 2025

Deutsche Börse AG

Data Analytics Working Student, Cash Market Services

Eschborn, Germany

Aug 2022 - Jul 2024

Arcadis

Data & Analytics Consultant, BIM Analytics

Bangalore, India

Feb 2022 - Jul 2022

Alliant Advisory

Associate, Cost Segregation

Bangalore, India

Jan 2021 - Jan 2022

Atal Incubation Centre - Bihar Vidyapith

Program Associate Intern, Entrepreneurship Cell

Patna, India

Education

2024 - 2026

Frankfurt School of Finance & Management

M.Sc. in Management · Frankfurt, Germany

2017 - 2021

Bangalore Institute of Technology

B.E. Civil Engineering · Bangalore, India

Professional credentials

Microsoft PL-300Power BI Data Analyst AssociateMicrosoft DP-600Fabric Analytics Engineer AssociateDatabricks DEAData Engineer AssociateMicrosoft AZ-900Azure FundamentalsMicrosoft AB-730AI Business Professional
Bloomberg Market ConceptsFinancial markets foundation · Available on request
McKinsey ForwardProblem solving and business skills

Open to Data & BI roles

Have a data problem
worth solving?

I’d like to hear about your team and what you’re building.

inbox.amitkumar18@gmail.com

Frankfurt am Main, Germany

LinkedIn GitHub