Senior Associate- Data Science (Pharma/Life Sciences/Healthcare domain

Chryselys; Hyderabad; Hybrid; INR 1,200,000 - 2,200,000; Full time

Job summary

Chryselys in Hyderabad, India, seeks a Data Scientist to collaborate with client stakeholders and analytics experts to develop predictive models and optimize operations.

You will work on experimental design, causal analysis, and data-driven solutions addressing business challenges. Strong collaboration and communication are valued for success in cross-functional teams.

Qualifications

• Experience with A/B testing and experiments design.
• Knowledge of sample size and power calculations.
• Familiarity with randomized control trials and bias avoidance.
• Understanding of quasi-experiment methods and causal inference.

Responsibilities

• A/B test sample size evaluation and variant assignment.
• Power analysis to detect meaningful effects.
• Design randomized controls to avoid bias between Test and Control variants.
• Apply quasi-experimentation methods like Diff-in-Diff or Propensity Score Matching.
• Differentiate between correlation and causality in models.

Skills

A/B test design

Power analysis

Experimental design

Randomized controls

Quasi-experimentation

Difference-in-Differences

Propensity score matching

Correlation vs causality

Job description

Role Overview:

As a Data Scientist at Chryselys, You will work closely with cross-functional teams, including client stakeholders, business consultants, and analytics experts, to develop predictive models, optimize operations, and deliver innovative solutions that address critical business challenges.

Responsibilities

Traditional Data science

• A/B test sample size evaluation, Test and control variant assignment, Quasi experimentation approach, Correlation models.


• Statistical Power Analysis: Experience in calculating sample sizes to ensure experiments have enough \\"\"power\\\"\" to detect meaningful effects.


• Experimental Design: Mastery of Randomized Controls, specifically how to assign \\"\"Test\\\"\" and \\"\"Control\\\"\" variants without introducing bias.


• Quasi-Experimentation: Proficiency in methods like Difference-in-Differences (Diff-in-Diff) or Propensity Score Matching.


• Correlation vs. Causality: Expertise in building models that distinguish between variables that move together and those that are causal.