Financial Markets Data · Engineering · London

Dhruv Verma

I understand financial markets well enough
to know when the data is wrong — and I engineer
the systems that keep it right.

LinkedIn GitHub Email

About

2016 B.Tech ECE
2020 MSc · Distinction
2023 FoodTech · Data Science
2024 Data & Operations · ETRM
Now
Financial Markets & Domain

Financial data problems are rarely just technical.

A broken risk report, a P&L that doesn't reconcile, a feed that looks clean but isn't — these trace back to schema changes, venue-side delivery shifts, or business rule drift that only someone with market context can diagnose correctly. My grounding is in European energy markets: ICE, EEX, Nord Pool, EPEX, LME/CME — I know how these feeds behave, where they break, and what correct looks like. That same diagnostic fluency applies wherever financial data quality has direct operational consequence.

Data Engineering

Python and SQL are my primary tools — I work at the full depth of the data stack.

I design and maintain ETL pipelines, feed validation workflows, stored procedure performance, and monitoring systems built for high-frequency datasets. I work at the systems level: tracing failures to root cause rather than patching symptoms, and building for production reliability from the start, not as an afterthought.

Applied AI & Automation

Where automation or AI tooling genuinely reduces operational overhead, I build it.

I'm comfortable across the Azure AI stack — Azure AI Search, OpenAI integration, App Service deployment, Bot Framework — and have used these in production. The judgement of when to reach for these tools matters as much as knowing how to use them.

Background

I work at the intersection of financial data systems and engineering — picking up the domain complexity a problem requires, and building the infrastructure to solve it properly.

My background spans ETRM platforms and energy market operations, financial data pipelines and warehouse architecture, commercial analytics and forecasting, and production AI systems. The common thread isn't a single vertical — it's the ability to move fluidly between market domain, data engineering, and applied tooling, and know which layer the problem actually lives in.

I've ingested and transformed live market feeds — ICE, EEX, Nord Pool, EPEX, LME/CME — built ETL pipelines and validation systems for high-frequency financial datasets, engineered financial reconciliation workflows across multi-market operations, designed revenue forecasting models from raw multi-stream data, and deployed AI tooling on Azure that reduced operational overhead in production. The problems have been different each time. The discipline has been the same: trace to root cause, build for reliability, don't patch symptoms.

I'm most useful where the data is complex, the domain is specialised, and the margin for error is low.

Built

Selected work

Operational AI Assistant

The Problem

Enterprise support workflows for a complex financial data platform generate years of historical issue data — patterns that are invisible when buried in a ticket backlog.

The Build

End-to-end AI assistant on Azure. Historical support tickets indexed using OpenAI embeddings and Azure AI Search with hybrid retrieval. GPT-4o surfaces contextualised fix suggestions for each query. Deployed via Bot Framework SDK on Azure App Service and surfaced through MS Teams — zero change to existing workflow.

Azure · Azure AI Search · OpenAI GPT-4o · RAG · Bot Framework SDK · Python · MS Teams

Multilevel Revenue Forecasting

The Problem

Multi-stream revenue data from partner markets needed accurate forward forecasts to support commercial planning decisions.

The Build

Time-series forecasting pipeline in NeuralProphet integrating multiple sales streams. Data cleaning, transformation, and feature engineering stages designed to maintain model accuracy across variable input quality. Results surfaced through operational dashboards.

NeuralProphet · Python · Pandas · AWS · Time-series forecasting · Dashboard reporting

Capabilities

Domain & execution

Domain Knowledge

  • Financial Markets Data
  • ETRM Systems & Product
  • Energy Market Feeds ICE · EEX · Nord Pool · EPEX · LME/CME
  • P&L & Risk Reporting
  • Feed Validation & Monitoring
  • ETL Pipeline Design
  • Financial Data Reconciliation
  • Data Warehouse Architecture
  • Root Cause Diagnostics

Technical Execution

  • Python — Pandas, NumPy, PySpark
  • SQL — PostgreSQL, T-SQL
  • C#
  • Azure — AI Search, Data Services, App Service, OpenAI Integration
  • AWS — Redshift, S3, RDS, QuickSight
  • Docker · Git
  • API Integration & Automation
  • Time-Series Forecasting
  • Dashboard & Reporting
MSc Big Data Science with Machine Learning Systems — Distinction
Queen Mary University of London · 2022–2023
B.Tech Electronics & Communication Engineering — 1st Class
Guru Gobind Singh Indraprastha University · 2016–2020
05 · Contact

Get in touch

Open to senior roles at the intersection of financial data engineering and domain complexity — energy, commodities, fintech, or wherever the data problem is hard.

Email me LinkedIn GitHub