About TWID
TWID is an AI incentive commerce platform that helps banks, UPI platforms and merchants influence customer decisions at checkout. Businesses spend billions on offers and reward points, yet most of that value goes unused or reaches the wrong customer. TWID uses AI to decide who should get an incentive, when and how much.
Our products, Nexus AI, Pay with Rewards and Checkout Intelligence, help partners acquire customers, grow spend and drive repeat purchases. TWID processes over a billion monthly transaction signals across 5,000+ merchants and 30+ issuers, including Swiggy, JioMart, ixigo and CRED.
TWID is backed by leading global investors, including Peak XV's Surge, Beenext, Rakuten and Google.
About the Role
We are looking for a Data Scientist with strong foundations in machine learning, experimentation, and data analytics, who can take models from development to production and translate business problems into measurable ML solutions.
This role will work closely with data and business problems emerging from TWID's transaction intelligence and incentive commerce ecosystem.
Key Responsibilities
2–3 years of hands-on experience building, evaluating, and deploying ML models that have successfully shipped to production.
Strong Python and SQL; independently explore large/complex datasets and feel comfortable being handed a data warehouse and finding your own way around it.
Strong foundations in classical ML: tree-based models, regression, classification, clustering, feature engineering, and model evaluation.
Strong rigor in experimentation and model evaluation: experiment design, appropriate metrics, statistical significance, and interpreting results.
Comfort with cloud data platforms, particularly AWS and its data/ML ecosystem.
Quantitative degree in Computer Science, Statistics, Mathematics, Engineering, Economics, a related field, or equivalent practical depth.
Strong written and verbal English communication, with the ability to explain technical concepts to both technical and non-technical stakeholders.
Nice to Have
FinTech AI/ML applications — risk, fraud, credit, customer analytics, and financial decisioning.
AI/ML-driven campaign management — segmentation, targeting, and optimization.
Causal inference and uplift modeling.
Recommender systems — collaborative filtering, sequential recommendation, embeddings, and deep learning.
LLM evaluation / Generative AI — hands-on experience with evaluation frameworks, prompt/model evaluation, RAG, agents, and the modern GenAI stack.