ZettaMine - Generative AI Engineer - LLM

About the Role :

As a Generative AI Engineer, you will play a key role in building, deploying, and scaling Generative AI-powered applications and services. You will work with state-of-the-art transformer models, build LLM pipelines, conduct fine-tuning, and leverage embedding techniques to develop intelligent applications in domains such as enterprise automation, document summarization, conversational AI, intelligent search, and more.

You will collaborate closely with data scientists, ML engineers, product teams, and cloud DevOps to bring GenAI-powered solutions from idea to production.

Key Responsibilities :

- Design and implement custom LLM pipelines using frameworks such as Hugging Face Transformers, LangChain, LlamaIndex, etc.

- Perform fine-tuning and RLHF (Reinforcement Learning with Human Feedback) on open-source or proprietary foundation models.

- Work with text generation, summarization, question-answering, chatbots, and agent-based workflows.

- Develop effective prompts and few-shot examples to enhance LLM performance.

- Design robust prompting strategies for instruction-following, zero-shot, and chain-of-thought reasoning.

- Optimize prompts for various LLM providers such as OpenAI, Anthropic, Cohere, or open-source models (e.g., Mistral, LLaMA, Falcon).

- Build and maintain vector-based search systems using FAISS, Pinecone, Weaviate, or similar tools.

- Integrate LLMs with RAG (Retrieval-Augmented Generation) pipelines to fetch contextually relevant data from custom knowledge bases.

- Work with unstructured text datasets, conduct preprocessing, and manage chunking, embeddings, and indexing strategies.

- Deploy models using containerized workflows (Docker, Kubernetes) and integrate them with cloud-native services (AWS/GCP/Azure).

- Automate model deployment pipelines using MLOps tools (MLflow, SageMaker, Vertex AI, etc.).

- Monitor LLM-based apps for latency, accuracy, hallucination risks, and cost optimization.

- Work with product managers, designers, and data teams to align on feature goals and delivery timelines.

- Collaborate with DevOps to set up secure, scalable APIs and infrastructure for LLM applications.

- Participate in code reviews, design discussions, and innovation brainstorming sessions.

Required Qualifications :

- Education: B.Tech/M.Tech/M.Sc in Computer Science, AI/ML, Data Science, or related fields.

- Experience: 510 years overall, with at least 34 years in Generative AI/LLMs.

- Proficiency in Python, strong coding practices, and understanding of software design principles.

- Deep hands-on experience with deep learning frameworks: PyTorch, TensorFlow.

- Experience with Hugging Face Transformers, LangChain, OpenAI API, vector databases, and embedding techniques.

- Solid understanding of LLM architecture, training, fine-tuning, and evaluation metrics.

- Experience with cloud platforms: AWS (e.g., Bedrock, SageMaker), GCP (Vertex AI), or Azure (OpenAI, ML Studio).

- Exposure to CI/CD for ML, model versioning, and inference optimization.