Data Scientist

Role Overview

As a Data Scientist at SquadStack.ai, you will play a crucial role in developing a lean, production-ready personalization layer for our outbound voice agent. Leveraging India's leading Voice AI platform's vast experience and data, you will contribute to building conversational superintelligence. This self-learning system aims to elevate end-consumer experiences through advanced personalization and contextualization, leading to increased conversions, greater satisfaction, and enhanced ROI for our clients.

Key Responsibilities

• Analyze real conversational data to understand and model natural and effective interaction elements.
• Experiment with machine learning techniques, translating insights into product enhancements and robust measurement frameworks while managing costs and performance.
• Collaborate closely with a dynamic team of product leaders, Field Data Engineers (FDEs), data scientists, and engineers to rapidly implement improvements.
• Continuously enhance conversation quality and outcome conversion rates through iterative development and feedback mechanisms.
• Take a hands-on approach to making voice AI agents more intelligent, adaptive, effective, and human-like.
• Engage in rapid learning and iteration with guidance from experienced advisors.

This is a hands-on, high-growth opportunity for a versatile data science professional skilled in transforming ambiguous challenges into measurable solutions with a strong focus on cost efficiency.

Required Qualifications

Experience:

• A minimum of 2 years of applied data science experience with a solid understanding of classical machine learning algorithms, including regression, classification, tree-based models, clustering, and foundational representation learning.
• Proficiency in hands-on supervised and unsupervised learning, feature engineering, and model validation techniques.

Experimentation:

• Basic knowledge of A/B testing, ROI framing, exposure/guardrail metrics, and familiarity with variance reduction techniques like CUPED.
• Working knowledge of causal inference concepts (e.g., propensity, uplift) and an understanding of how to avoid common pitfalls.

Ownership and Collaboration:

• Demonstrated ability to independently research, scope, build, and evaluate solutions from data acquisition to production deployment.
• Strong collaboration skills to work effectively with ML engineers, product managers, operations teams, and FDEs to ensure impact and quality.

Technical Skills:

• Proficiency in Python, including libraries such as scikit-learn, pandas, and NumPy.
• Experience with gradient boosting libraries like XGBoost or LightGBM.
• Proven experience shipping models to production and establishing monitoring systems.
• Basic familiarity with AWS services (EC2, S3), with SageMaker being a plus.
• Understanding of MLOps trade-offs, including deployment, monitoring, versioning, and the differences between streaming and batch inference.

Preferred Qualifications

• Experience implementing personalization or recommendation engines, particularly in sales, marketing, or lead-oriented B2C environments.
• Exposure to reinforcement learning or bandit algorithms, with the ability to implement simple models like LinUCB/LinTS under guidance.
• Experience with graph-based modeling or Graph Neural Networks (GNNs) for lead intelligence or lookalike feature development.
• Experience building dashboards and delivering business-facing analytics.
• Experience with PyTorch for developing simple sentiment analysis or embedding models.
• Familiarity with Text-to-Speech (TTS) or Automatic Speech Recognition (ASR) technologies is a plus.

Logistics

• Compensation: Competitive, based on experience and qualifications.
• Joining: As soon as possible.
• Location: Noida or Goa, India.