Lead Data Scientist - NLP/Artificial Intelligence

Description :

Lead Data Scientist - AI/ML/NLP

- 69 years

Domain ML Leadership Applied Modelling Experimentation Team Mentorship

The Role :

Lead Data Scientists own the ML function for a defined product domain taking business questions from ambiguity through model training through A/B validation and into production.

You are the first person the product team calls when a metric breaks, the engineer who can explain to a sceptical stakeholder why a model made a specific prediction, and the mentor who raises the quality of the scientists around you.

- Own ML for a product domain end-to-end from OKR translation to production deployment

- Lead a pod of 23 Data Scientists; your code reviews are learning experiences, not gatekeeping

- Drive experimentation culture: every change is a hypothesis, every hypothesis gets a test

- Your models are trusted by the product team because your evaluation is airtight

Core Responsibilities :

Model Ownership Across Platforms :

- Formulate ML problems from product briefs: translate 'improve hotel search' into a ranking objective with measurable proxy and business metrics

- Build, evaluate, and maintain ML models for your domain: ranking, recommendation, price prediction, NLP classification, or fraud scoring

- Own feature engineering for your domain: design features, build pipelines, document semantics, monitor for drift

- Manage model debt: schedule retraining, track performance over time, sunset underperforming models before they cause incidents

AI-First Feature Delivery :

Ranking & Relevance: train and deploy L2R models for your vertical; run position bias experiments; evaluate with both offline and online metrics

Price Intelligence: build price sensitivity models per user segment; deliver fare prediction features for the recommendation feed

Recommendation: implement collaborative filtering or content-based models for cross-sell and upsell surfaces in your domain

LLM/RAG contribution: own a specific RAG pipeline (e.g., hotel cancellation policy Q&A); build evaluation datasets and monitor retrieval quality

Sentiment & Reviews: train review summarisation models; build complaint classification systems for support routing in your vertical

Voice AI contribution: build intent classification models for spoken queries in your domain; evaluate ASR output quality on domain vocabulary

Coupon ML: build personalised discount targeting models; measure redemption lift vs. margin cost with causal evaluation

Experimentation & Analytics :

Design A/B tests for ML model changes: power analysis, SRM detection, guardrail metric selection, and post-experiment diagnostics

Build domain-specific evaluation frameworks: what does 'good' look like for hotel ranking? for fare prediction? you define and defend the answer

Produce experiment read-outs that non-technical stakeholders trust and act on

The AI-First Mandate :

- AI is not an enhancement.

- It is the product architecture.

- Every surface, every API, every decision point is either ML-powered today or on the roadmap to be.

- Search & Ranking Learning-to-Rank across flights, hotels, bus routes, train coaches; real-time re-ranking on user signals

- Voice AI Hindi/Hinglish voice booking, intent resolution, spoken fare comparisons, accessibility-first conversational UX

- RAG Systems - Fare rule retrieval, hotel cancellation policy Q&A, airline contract intelligence, real-time regulatory updates

- Agentic AI - Autonomous booking resolution, exception handling, refund orchestration, supplier communication bots

- MCP Orchestration - Model Context Protocol tool chains across GDS APIs, payment gateways, and supplier integrations

- Recommendation Engine - Cross-vertical next-best-action, collaborative filtering, session-based deep learning

- Price Intelligence - Competitive fare mapping, lower-price guarantee engine, demand elasticity, yield optimisation

- Coupon & Promo ML - Personalised offer targeting, redemption probability scoring, margin-aware discount optimisation

- Sentiment & Review AI - Review summarisation, NPS prediction, complaint triage, trust signal extraction

- Fraud & Risk ML - Anomaly detection, account takeover signals, payment fraud scoring, fake review classification

- Deep System Mapping - Route intelligence, geo-semantic matching, multi-modal journey planning

- Predictive Systems - Cancellation risk, no-show prediction, seat upgrade probability, waitlist conversion

Who You Are :

- 69 years in Data Science/ML with at least 2 end-to-end production deployments from problem definition through monitoring

- Strong in ML fundamentals: you can train a ranking model, run a causal experiment, and explain the result to an engineering director

- Working NLP experience: text classification, embeddings, or fine-tuning LLM experience is a strong plus

- Comfortable leading a small technical pod: delegating work, giving feedback, and holding the quality bar without micromanaging

- Tier-I institute preferred (IIT / IIIT / NIT / IISC / BITS CSE / Statistics / AI)

Technology Stack :

ML/DL : Scikit-learn PyTorch XGBoost LightGBM CatBoost

NLP/LLM : HuggingFace Transformers spaCy LangChain Sentence Transformers

Data Eng : Python SQL Spark Airflow dbt Feast

Serving & Exp : FastAPI MLflow Docker Weights & Biases