Roles & Responsibilities :

- Partner with Product to spot high-leverage ML opportunities tied to business metrics.
- Wrangle large structured and unstructured datasets; build reliable features and data contracts.
- Build and ship models to :

i. Enhance customer experiences and personalization ii. Boost revenue via pricing/discount optimization iii. Power user-to-user discovery and ranking (matchmaking at scale)

iv. Detect and block fraud/risk in real time v. Score conversion/churn/acceptance propensity for targeted actions.

- Collaborate with Engineering to productionize via APIs/CI/CD/Docker on AWS.
- Design and run A/B tests with guardrails.
- Build monitoring for model/data drift and business KPIs.

Ideal Candidate :

- Profile : Strong Data Scientist/Machine Learning/AI Engineer Profile who trains models on first-party data. (Not a GenAI / LLM application engineer)
- Mandatory (Experience 1) : Must have 3 years of hands-on experience as a Data Scientist or Machine Learning Engineer building ML models with atleast the recent 1 years into trained-model work, with a product company.
- Mandatory (Experience 2) : Must have robust expertise in Python with the ability to implement classical ML algorithms including linear regression, logistic regression, decision trees, gradient boosting, etc. Must be evidenced by a named model on a named dataset, not by a skills-section listing.
- Mandatory (Experience 3) :



Must have practical hands-on experience with Neural Network / Deep Learning models (TensorFlow / PyTorch preferred) that the candidate trained or fine-tuned themselves. LoRA / PEFT / SFT fine-tuning counts.
- Mandatory (Experience 4) : Must have hands-on experience in minimum 2 use cases out of recommendation systems, image data, fraud/risk detection, price modelling, propensity models each owned end to end (dataset, features, training, evaluation, deployment).
- Mandatory (Experience 5) : Must have strong hands on exposure to NLP, on the representation side : text classification, embeddings, similarity models, user profiling, NER and feature extraction from unstructured text.
- Mandatory (Experience 6) : Must have experience productionising ML models through APIs/CI/CD/Docker and working on AWS or GCP environments.
- Mandatory (Experience 7) : Every ML claim must carry a stated evaluation metric with a value (ROC-AUC, FPR, accuracy, precision/recall, MRR, precision@k, NDCG, RMSE) or a measured business outcome. A resume that lists algorithms and tools with no numbers attached is a reject.
- Mandatory (Company) : Must be from product companies, Avoid candidates from financial domains (e.g., JPMorgan, banks, fintech).
- Mandatory (Stability) : Must have stayed for a minimum of 2 years with each of the previous companies. Short tenures are acceptable provided the candidate has demonstrated stability through longer tenures elsewhere in their career.

📌 Data Scientist - Machine Learning/Artificial Intelligence (India)
🏢 The SEO Byte
📍 India