Lead Data Scientist - Generative AI

Job Summary :

We are looking for an experienced AI - Generative AI Developer who can work across the AI spectrum - from classical machine learning models to cutting-edge Generative AI applications.

The role demands strong experience in building ML models using regression, classification, and tree-based algorithms, along with hands-on exposure to LLMs and generative frameworks like GPT, Stable Diffusion, and LangChain.

Classical AI/ML Responsibilities :

• Design and implement supervised and unsupervised ML models including Linear Regression, Logistic Regression, Decision Trees, Random Forest, XGBoost, Naive Bayes, K-Means, SVM, PCA, etc.

• Preprocess and analyse structured/tabular datasets.

• Evaluate models using metrics like accuracy, precision, recall, ROC-AUC, and RMSE.

• Build predictive models, deploy them into production, and monitor performance.

• Collaborate with business teams to translate requirements into ML use cases.

Generative AI (GenAI) Responsibilities :

• Build and fine-tune LLMs (e.g., GPT, LLaMA, PaLM) for summarisation, Q&A, document generation, etc.

• Implement prompt engineering, RAG pipelines, and vector database integrations.

• Use libraries like Hugging Face Transformers, LangChain, and LlamaIndex.

• Develop APIs to expose GenAI models in real-time apps.

• Optimise model inference using quantisation, batching, etc.

• Ensure safe, explainable, and bias-free output in alignment with AI ethics guidelines.

Qualifications :

• Bachelor's or Master's in Computer Science, Data Science, Statistics, or related field.

• Strong programming skills in Python, with experience in NumPy, Pandas, Scikit-learn.

• Proficiency in classical ML algorithms (regression, trees, naive Bayes, etc.).

• Experience with LLM frameworks like OpenAI API, Hugging Face, and LangChain.

• Understanding of transformer architecture, NLP, embeddings, and tokenisation.

• Familiarity with REST API development using FastAPI/Flask.

• Exposure to cloud platforms (AWS/GCP/Azure) and Docker/Kubernetes.

Preferred Skills :

• Experience with deep learning (TensorFlow, PyTorch).

• Exposure to image/audio/video generation using models like DALL-E, Stable Diffusion, Whisper.

• Familiarity with RAG, LLMOps, and vector stores (FAISS, Pinecone, Weaviate).

• Knowledge of MLOps pipelines, model monitoring, and CI/CD for ML.