Quant Developer

Value Research; Dadri; On-site; INR 900,000 - 1,500,000; Full time

Benefits offered by this job

Ownership of data tooling

Shape how things are built

Exposure to rating methodology

Collaboration with Quant Desk

Job summary

Value Research in India seeks a quant data engineer to automate data workflows, build Python pipelines over large financial datasets, and ensure traceable source data for research and ratings.

You will collaborate with analysts to stress-test models, implement automated data quality checks, and develop extraction from company documents, with emphasis on reliability and clear communication.

Qualifications

• Graduate in engineering, computer science, statistics, mathematics, economics, finance, or any discipline paired with real ability in code and numbers.
• Python that runs unattended, not just in a notebook: pandas and NumPy, plus error handling, logging and retries.
• SQL and working comfort with a relational database. PostgreSQL preferred.
• Statistics understood rather than recited: dispersion, distributions, regression, what R-squared and a p-value actually tell you, and why a 15 per cent average return can hide a great deal.
• Understands overfitting and data leakage, and knows why validating a time series is not the same as validating a cross-section.
• Can read a profit and loss statement, a balance sheet and a cash flow statement, or the appetite to become able to quickly.
• Verifies every number to its source, and treats a silent failure as the worst kind.
• Can explain technical work plainly to an analyst who does not code.

Responsibilities

• Automate the data work analysts currently do by hand, so that retrieval, cleaning and loading run to a schedule logged, monitored, and recoverable when something breaks.
• Build and maintain Python-based analytical pipelines over large financial datasets, with the source of every value traceable.
• Work with analysts to implement, refactor and stress-test our quantitative models, scores and rating frameworks.
• Build and run backtests properly: point-in-time data, appropriate validation, no look-ahead. Explain what a result does and does not establish.
• Build automated data quality checks and monitoring, so problems surface to a person before they reach a reader.
• Develop reliable extraction of financial data from published company documents, and the review process for what automated extraction gets wrong.
• Support the equity data team with the tooling they need to run their operations without depending on you day to day.
• Give analysts early, honest feedback on method where a specification is ambiguous, where a result is fragile, where the data cannot support the claim being made.

Skills

Python

SQL

Statistics

Pandas

NumPy

Education

Engineering / CS / Stats / Math / Economics / Finance

Tools

PostgreSQL

statsmodels

scikit-learn

Grafana

Looker Studio

Git

Testing

Job description

Role overview

The quant team builds the scoring frameworks, ratings and research tooling that sit underneath our stock research and our advisory services. Analysts own the investment logic. You own the machinery that makes it run the data that feeds it, the code that computes it, the tests that prove it works, and the checks that catch it when it does not.

This is not a pure engineering role. You will be asked what you think. If a model's output looks wrong, or a backtest looks too good, we expect you to say so and to be able to explain why. That means understanding the statistics well enough to hold an opinion, not just enough to call the function.

One thing to be clear about up front. We are long-term fundamental investors. The quant work here is about measuring business quality, financial health and valuation over years. It is not signal-chasing, not high frequency, and not a search for edges that decay in a week. If that is the work you want, this is the wrong desk.

Key responsibilities

• Automate the data work analysts currently do by hand, so that retrieval, cleaning and loading run to a schedule logged, monitored, and recoverable when something breaks.
• Build and maintain Python-based analytical pipelines over large financial datasets, with the source of every value traceable.
• Work with analysts to implement, refactor and stress-test our quantitative models, scores and rating frameworks.
• Build and run backtests properly: point-in-time data, appropriate validation, no look-ahead. Explain what a result does and does not establish.
• Build automated data quality checks and monitoring, so problems surface to a person before they reach a reader.
• Develop reliable extraction of financial data from published company documents, and the review process for what automated extraction gets wrong.
• Support the equity data team with the tooling they need to run their operations without depending on you day to day.
• Give analysts early, honest feedback on method where a specification is ambiguous, where a result is fragile, where the data cannot support the claim being made.

Qualification/Experience

• Graduate in engineering, computer science, statistics, mathematics, economics, finance, or any discipline paired with real ability in code and numbers.
• Python that runs unattended, not just in a notebook: pandas and NumPy, plus error handling, logging and retries.
• SQL and working comfort with a relational database. PostgreSQL preferred.
• Statistics understood rather than recited: dispersion, distributions, regression, what R-squared and a p-value actually tell you, and why a 15 per cent average return can hide a great deal.
• Understands overfitting and data leakage, and knows why validating a time series is not the same as validating a cross-section.
• Can read a profit and loss statement, a balance sheet and a cash flow statement, or the appetite to become able to quickly.
• Verifies every number to its source, and treats a silent failure as the worst kind.
• Can explain technical work plainly to an analyst who does not code.

Good to have

• Extracting structured data from documents at scale, including older and inconsistent formats.
• statsmodels, scikit-learn, or equivalent modelling experience.
• Orchestration and workflow tooling, and dashboarding in Grafana, Looker Studio or similar.
• Git, code review, and writing tests as a habit rather than an instruction.
• Exposure to the Indian market and company data, and the ways it is routinely messy.
• Using LLMs for extraction, together with the judgment to know where their output must not be trusted.
• NISM Series XV, Research Analyst certification, or progress towards CFA or CA. We will support you in obtaining certification.

What we offer

• Ownership of the data and quant tooling layer behind research that reaches a very large retail audience.
• Scope to shape how things are built, rather than maintaining someone else's design.
• Close work with a research desk that has been at this since 1990, and direct exposure to rating methodology built over three decades.
• Real say in method. Analysts here argue with each other, and you are expected to join in.
• Competitive compensation, based on ability rather than years served.

Work from Office: Monday to Friday

Saturdays & Sundays Fixed Off