Project Role : Data Science Practitioner
Project Role Description : Formulating, design and deliver AI/ML-based decision-making frameworks and models for business outcomes. Measure and justify AI/ML based solution values.
Must have skills : Data Science
Good to have skills : NA
Minimum 5 year(s) of experience is required
Educational Qualification : 15 years full time education

Summary: Design, develop, and deploy Python-based applications leveraging Generative AI models, LLMs, and AI-powered APIs for enterprise and product use cases. Roles & Responsibilities: - Develop AI-powered applications integrating LLMs (OpenAI GPT, Claude, LLaMA, Mistral, etc.) via APIs and SDKs - Fine-tune and deploy custom models using Hugging Face Transformers, LangChain, or Haystack - Implement prompt engineering and retrieval-augmented generation (RAG) pipelines - Build vector search integrations with Pinecone, Weaviate, FAISS, Milvus - Design ingestion pipelines for structured/unstructured data using Pandas, PySpark, or Dask - Integrate AI features into backend APIs using FastAPI, Flask, or Django - Implement embeddings generation, semantic search, and context injection - Optimize model serving with ONNX, TensorRT, or quantization techniques - Deploy on cloud services (AWS Sagemaker, Azure ML, GCP Vertex AI) or containerized environments - Build CI/CD pipelines for AI apps with GitHub Actions, Jenkins, or Azure DevOps Professional & Technical Skills: - Python 3.8+ (async programming, OOP, functional patterns)• LLM API integration (OpenAI, Anthropic, Cohere, Mistral, Azure OpenAI)• Prompt engineering & RAG architectures• LangChain, LlamaIndex, Haystack frameworks• Vector databases (Pinecone, Weaviate, FAISS, Milvus)• Data preprocessing for NLP – tokenization, cleaning, embedding generation• Model fine-tuning (Hugging Face Trainer, LoRA, PEFT)• Async API development for AI endpoints• Performance optimization for inference latency - Experience with multimodal models (image, audio, video)• Knowledge of RLHF (Reinforcement Learning with Human Feedback)• Knowledge of GPU optimization with CUDA, PyTorch Lightning• Familiarity with Kubernetes for scalable AI deployment• Integration with enterprise systems for AI-assisted workflows• Prompt engineering - Python, Hugging Face, LangChain, LlamaIndex, Haystack, PyTorch, TensorFlow, ONNX, Docker, Kubernetes, AWS Sagemaker, Azure ML, GCP Vertex AI, Git, GitHub/GitLab, Postman, Swagger/OpenAPI Additional Information: - The candidate should have minimum 5 years of experience in Data Science. - This position is based at our Pune office. - A 15 years full time education is required.