Quantitative Analyst

Quantitative Analyst
Location: Ahmedabad (on-site)
Type: Full-time
Experience: 1-5 years
_________________________________________________
About the Role
We are a proprietary quantitative trading firm focused on Indian and global equity, derivatives, and cryptocurrency
markets. This role is responsible for researching trading ideas, analysing market data and developing systematic
strategies, with scope to investigate other markets, instruments and asset classes where relevant.
You will take research from an initial observation through to a conclusion: forming a hypothesis about market
behaviour, preparing the data required to test it, constructing signals and rules, building and running the backtest,
and determining whether the result holds up under further scrutiny. You will work closely with the team responsible
for data and trading systems to specify what your research needs and to help suitable strategies progress towards
live implementation.
Python is central to this role. You will write your own research and backtesting code and must be able to inspect and
debug it yourself. Meaningful knowledge of at least one market - Indian, US or cryptocurrency - is required, along
with the ability to investigate unfamiliar markets and instruments carefully. This is not an entry-level position.
Responsibilities
• Develop and test research ideas. Form research questions from market observations, economic reasoning,
statistical patterns or observed strategy behaviour. Be able to explain why an effect might exist, and what
would have to be true for it to persist, before committing to testing it.
• Design strategies end to end. Define instruments, universe, observation frequency, signals, entry and exit
conditions, filters, holding period, position sizing and risk controls. Explain the reasoning behind the universe,
including instruments deliberately excluded and why.
• Investigate market and instrument mechanics. Understand how trading sessions, liquidity, contract
specifications, costs, order behaviour and execution constraints affect whether a strategy is feasible in practice
rather than only on paper.
• Write research and backtesting code in Python. Prepare data, run exploratory analysis, calculate signals,
implement backtests and evaluate results, working extensively in pandas and NumPy. Write, inspect and debug
this code independently.
• Work carefully with financial time-series data. Handle identifiers, timestamps, timezones, differing frequencies
and datasets from multiple sources. Investigate missing observations, inconsistent prices, corporate-action
adjustments and other data issues before relying on any result.
• Evaluate results critically. Examine performance across time periods, instruments and market conditions.
Investigate drawdowns, turnover, concentration, trade counts, parameter sensitivity and the effect of realistic
trading costs and assumptions. Distinguish a genuine effect from one produced by the testing process itself.
• Keep exploration separate from evaluation. Reserve data that has not been used to select the model or its
parameters, and be able to state what a given result does and does not demonstrate.
• Take promising strategies towards implementation. Specify data requirements, validate signals and
calculations against the production data layer, and support the work required to move a strategy into the live
framework.
• Analyse live and paper performance. Where a strategy has been deployed, compare its behaviour against the
backtest and help establish whether differences arise from data issues, implementation differences, execution
effects or a change in the underlying opportunity.

• Maintain clear records of your research. Keep hypotheses, datasets used, code, results, attempted variations
and rejected approaches organised, so that conclusions can be reproduced and re-examined.
What We Expect
• Relevant professional experience in quantitative research, systematic trading, strategy analysis or closely
related market work. You should already have built and evaluated strategies of your own.
• Practical understanding of a financial market. Indian, US or cryptocurrency. You should understand how the
market you have worked in actually functions - instruments, sessions, liquidity, costs, order behaviour - and
what the available data does and does not represent.
• Strong hands-on Python ability. Particularly pandas and NumPy. You should be comfortable writing research
code from scratch, working with time-series and panel data, and handling messy real-world datasets without
depending on a framework to structure the work for you.
• Working knowledge of statistics and probability. Returns and distributions, sampling uncertainty, correlation
and its limitations, regression, hypothesis testing, stationarity, sample size, and the common ways in which a
result can be fitted rather than real.
• Experience cleaning and analysing real financial datasets, with attention to timestamp alignment, data
availability, identifier consistency and historical accuracy.
• The ability to explain a strategy you have built in complete detail - what it captures, why you expected it to
work, what it assumes, how it was tested, where it performs poorly and what remains unresolved.
• Sound judgment about what to do next. Ability to discuss failures and limitations directly, identify which
additional checks matter, and decide which experiments are worth running.
• Ability to communicate clearly and work with colleagues on data, validation and implementation questions.
• Comfort working on Linux, reading and writing SQL, and working with large datasets.
Preferred Experience
• Taking a researched strategy through paper trading or live implementation, and investigating the differences
from its backtest.
• Experience across more than one market, and the ability to compare how an idea behaves in different market
structures.
• SQL and direct experience with data held in PostgreSQL, ClickHouse, Parquet or CSV.
• Obtaining or validating data through REST APIs or WebSocket feeds.
• Building a backtesting framework from scratch, rather than only using an existing one.
• Derivatives and options, cryptocurrency spot and perpetual instruments, market microstructure, or execution
and transaction-cost analysis.
• More advanced statistical or machine-learning methods, where you can explain why the method suited the
research problem.
What This Role Is Not
• A reporting or business-intelligence role. Dashboards, MIS, visualisation tools and Excel-based analysis are not
the deliverables.
• A discretionary trading role. This is systematic research. Trading decisions based primarily on personal
judgment, manual chart reading or reacting discretionarily to news are not the work.
• A sales, advisory or client-facing role. The position does not involve fundraising, client acquisition or providing
investment advice.
• A role that ends at the notebook. Research is expected to be reproducible, examined properly and taken far
enough that a conclusion can be relied upon.
• A tool-driven role. Research is implemented in Python. Platform-based or GUI backtesters may appear in your
background, but they are not a substitute for writing the analysis yourself.