Gen AI Developer - AWS Cloud

EY; Hyderabad; On-site; INR 3,600,000 - 6,000,000; Full time

Job summary

EY Hyderabad is seeking a Senior Data Scientist to lead research, design, and deployment of advanced generative AI models. You will own projects from ideation to production, applying deep ML/DL expertise to solve complex business problems.

You will work with a broad stack including PyTorch, Transformers, LangChain, and MLflow, driving delivery, monitoring, and performance in a fast-paced environment.

Qualifications

• Masters or Bachelor’s degree in Computer Science or related field.
• 3.5–7 years of AI/ML experience with end-to-end delivery.
• Hands-on experience productionizing ML/LLM solutions.
• Strong ML/DL theory foundations and analytical skills.
• Publications, open-source contributions, or patents are a plus.

Responsibilities

• Research and design advanced generative AI models.
• Implement production-ready AI solutions and pipelines.
• Own projects from ideation through production.
• Stay current with cutting-edge generative AI techniques.

Skills

Python

Deep learning

ML theory

Optimization

Problem solving

Education

Master's or Bachelor's in CS/related

Tools

PyTorch

TensorFlow

JAX

HuggingFace

LangChain

MLflow

Kubeflow

FAISS

CUDA

Job description

Role Overview Senior Data Scientist As a Senior Data Scientist specializing in AI, you will be a key individual contributor responsible for the research, design, and implementation of advanced generative AI models. You will tackle some of our most complex business challenges, transforming concepts into tangible, high-impact AI solutions. This role requires a deep technical background in machine learning, Traditional ML, Advance ML, Deep learning, Statistical modelling techniques and a passion for staying on the cutting edge of generative AI techniques, including LLMs, diffusion models, and agentic systems. You will own projects from ideation through to production, applying your expertise to build models that push the boundaries of what's possible.

Required Technical Expertise

• Languages & Frameworks
• Python (core ML/AI language, advanced data structures, async & multiprocessing)
• Deep learning frameworks: PyTorch, TensorFlow, JAX
• HuggingFace Transformers & Diffusers ecosystem
• LLM/agentic frameworks: LangChain, LangGraph, LlamaIndex, Semantic Kernel
• MLOps: MLflow, Weights & Biases, Kubeflow
• Graph Rag, MCP frameworks
• Working with CUDA libraries and optimization (CuGraph, CuPy)

Generative & Agentic AI

• Retrieval-Augmented Generation (RAG) standard, graph-based, and vector DBintegrated (FAISS, Pinecone, Weaviate, Milvus)
• Multi-agent orchestration (LangGraph, AutoGen, Semantic Kernal, tool-calling with OpenAI function APIs)
• Fine-tuning diffusion & generative models (Stable Diffusion, Flux.1, ControlNet, DreamBooth)
• LLM fine-tuning (parameter-efficient methods: LoRA, QLoRA, adapters) on models like Llama3, Mistral, CodeLlama, Falcon, Gemma, Azure OpenAI models
• Advanced prompt engineering with context engineering(system prompting, function calling, safety alignment, guardrails

Machine Learning & Deep Learning

• Classical ML: tree ensembles (XGBoost, LightGBM, CatBoost), linear/logistic regression, clustering (kmeans, DBSCAN), dimensionality reduction (PCA, t-SNE, UMAP)
• Deep learning: CNNs (ResNet, EfficientNet), RNNs/LSTMs, GRUs, Transformer architectures (BERT, ViT, GPT)
• Graph ML: Graph Neural Networks (GNNs GraphSAGE, GAT, PyTorch Geometric, DGL)

Time Series: forecasting (Prophet, ARIMA, DeepAR, Temporal Fusion Transformer)

• Reinforcement Learning: RLHF, PPO, DQN, policy gradients

Computer Vision

• Object detection: YOLO (v5v8), DETR, Faster R-CNN, SSD
• OCR: PaddleOCR, Tesseract, EasyOCR, LayoutLM for document understanding
• Video analytics: object tracking (DeepSORT, ByteTrack), frame stitching, camera calibration & SLAM pipelines
• Multi-modal ML: CLIP, BLIP, Florence-2, Segment Anything (SAM), vision-language grounding

Optimization & Deployment

• Model optimization: TensorRT, ONNX Runtime, quantization (INT8, FP16, mixed precision), pruning, distillation
• Scalable deployment: Dockerized microservices, Kubernetes, REST/gRPC APIs, Azure functions.
• Cloud: Azure AI/ML services (App Service, Azure OpenAI, AML)
• Monitoring: Azure Monitor, Prometheus, Grafana, ELK stack for model drift

Qualifications

Masters or Bachelor’s degree in Computer Science or related field

3.5–7 years of experience in AI/ML with demonstrated end-to-end solution delivery

Hands-on experience productionizing ML/LLM solutions (from PoC scalable deployment monitoring & maintenance)

Strong foundation in ML/DL theory (optimization, loss functions, architectures)

Demonstrated applied research/engineering — publications, open-source contributions, or patents are a plus

Strong analytical, problem-solving, and debugging skill