Data Scientist

Role Overview
In this role, you will help build a lean, production-ready personalization layer for our outbound voice agent.
As India's leading Voice AI platform, Squadstack has vast experience and data across human and AI voice agent calling. We are building conversational superintelligence - a self learning system to enhance the experience for the end consumers through hyper personalization and hyper contextualization, leading to higher conversions, satisfaction and improved RoI for our clients.
What you'll be doing
Dive into real conversational data to understand and model what makes interactions feel natural and effective.
Experiment with machine learning ideas and turn learnings into tangible product upgrades and measurement frameworks, while balancing cost and performance.
Partner closely with a small, fast-moving team of product leaders, FDEs, data scientists and engineers to ship improvements quickly.
Continuously raise the bar on conversation quality and outcome conversion through rapid iteration and feedback loops.
Play a hands-on role in making voice AI agents smarter, more adaptive, more effective and more human over time.
Learn and iterate rapidly with guidance from experienced advisors.
This is a hands-on high growth role for a data science generalist who can turn ambiguous problems into measurable solutions with strong cost discipline.
What We Want
Required
2+ years applied data science experience with strong foundations in classical ML:
Regression/classification, tree-based models, clustering, representation learning basics.
Hands-on supervised and unsupervised learning, feature engineering, model validation.
Experimentation basics:
A/B testing, ROI framing, exposure/guardrail metrics; familiarity with CUPED or other variance reduction.
Working knowledge of causal inference concepts (propensity, uplift) and how to avoid common pitfalls.
End-to-end ownership:
Ability to independently research, scope, build, and evaluate solutions from data sourcing to production rollout.
Ability to collaborate with other ML, Product, Ops and FDEs to ensure impact and quality controls
Technical/ Operational experience:
Proficient in Python, scikit-learn, pandas, Num Py; experience with XGBoost/Light

GBM.
Has shipped models to production and monitored them.
AWS basics (EC2, S3; Sage Maker a plus); understands MLOps tradeoffs (deployment, monitoring, versioning, streaming vs batch inference).
Preferred
Implementation of a personalization/ recommendation engine or related sales/marketing/ lead oriented solutions in a B2C environment.
Exposure to reinforcement learning or bandits; comfortable implementing simple  LinUCB/LinTS  with guidance.

Experience with  graph-based modeling/GNNs  for lead intelligence or lookalike features.
Experience building dashboards and business-facing analytics.
PyTorch  experience for simple sentiment/embedding models.
TTS/ASR  familiarity is a plus.
Logistics
Compensation:
Competitive!
Joining: ASAP!

Location:

Noida/Goa