Principle Developer - ML/Prompt Engineer

Shashwath Solution; Pune District; On-site; INR 1,500,000 - 2,100,000; Full time

Job summary

Shashwath Solution seeks a Principle Developer - ML/Prompt Engineer to design, deploy, and maintain a Retrieval Augmented Generation (RAG) model on Amazon Bedrock, leveraging cutting-edge LLMs and a unified API. You will craft prompts, optimize performance, and ensure security and reliability across GenAI applications.

You will collaborate across teams to implement scalable GenAI systems using Java, Python, C/C++, and AWS services, while refining prompts and workflows for varied use cases.

Qualifications

• Experience in programming with Java, Python, or C/C++.
• Experience with generative AI tools, models, and frameworks (Anthropic, OpenAI, Hugging Face, TensorFlow, PyTorch).
• Experience with RAG models or similar architectures (RAG, Ragna, or Pinecone).
• Experience with Bedrock or similar platforms (AWS Lambda, SageMaker, or Comprehend).
• Ability to design, iterate, and optimize prompts for various LLM use cases.
• Awareness of prompt engineering techniques (zero-shot, few-shot, chain-of-thought).
• Familiarity with prompt evaluation strategies, including metrics and A/B testing frameworks.
• Experience building prompt libraries and reusable templates for GenAI apps.
• Ability to debug and refine prompts for accuracy, safety, and alignment.
• Awareness of prompt injection risks and mitigation strategies.
• Familiarity with PEFT and prompt chaining methods.
• Familiarity with continuous deployment and DevOps tools; Git experience is a plus.

Responsibilities

• Develop, deploy, and maintain a RAG model in Amazon Bedrock.
• Design and implement a RAG model that can generate natural language responses and actions based on user queries.
• Integrate the RAG model with Bedrock and related APIs for scalable GenAI apps.
• Optimize the RAG model for performance, scalability, and reliability.
• Design, test, and optimize prompts for diverse LLM use cases (summarization, classification, translation, Q&A, agent workflows).
• Develop and maintain reusable prompt templates, chains, and libraries to support scalable GenAI applications.

Skills

TensorFlow

PyTorch

Java

Agile

Scrum

DevOps

CI/CD

Jupyter Notebook

AWS Lambda

C++

Git

AWS SageMaker

ML

Python

Natural Language Processing

Deep Learning

Data Science

Keras

Artificial Intelligence

Neural Networks

NLP

Tools

Amazon Bedrock

OpenAI

Hugging Face

TensorFlow

Jupyter

PyTorch

Job description

Principle Developer - ML/Prompt Engineer
Technologies: Amazon Bedrock, RAG Models, Java, Python, C or C++, AWS Lambda,

Responsibilities

Responsible for developing, deploying, and maintaining a Retrieval Augmented Generation (RAG) model in Amazon Bedrock, our cloud-based platform for building and scaling generative AI applications.
Design and implement a RAG model that can generate natural language responses, commands, and actions based on user queries and context, using the Anthropic Claude model as the backbone.
Integrate the RAG model with Amazon Bedrock, our platform that offers a choice of high-performing foundation models from leading AI companies and Amazon via a single API, along with a broad set of capabilities to build generative AI applications with security, privacy, and responsible AI.
Optimize the RAG model for performance, scalability, and reliability, using best practices and robust engineering methodologies.
Design, test, and optimize prompts to improve performance, accuracy, and alignment of large language models across diverse use cases.
Develop and maintain reusable prompt templates, chains, and libraries to support scalable and consistent GenAI applications.

Skills/Qualifications

Experience in programming with at least one software language, such as Java, Python, or C/C++.
Experience in working with generative AI tools, models, and frameworks, such as Anthropic, OpenAI, Hugging Face, TensorFlow, PyTorch, or Jupyter.
Experience in working with RAG models or similar architectures, such as RAG, Ragna, or Pinecone.
Experience in working with Amazon Bedrock or similar platforms, such as AWS Lambda, Amazon SageMaker, or Amazon Comprehend.
Ability to design, iterate, and optimize prompts for various LLM use cases (e.g., summarization, classification, translation, Q&A, and agent workflows).
Deep understanding of prompt engineering techniques (zero-shot, few-shot, chain-of-thought, etc.) and their effect on model behavior.
Familiarity with prompt evaluation strategies, including manual review, automatic metrics, and A/B testing frameworks.
Experience building prompt libraries, reusable templates, and structured prompt workflows for scalable GenAI applications.
Ability to debug and refine prompts to improve accuracy, safety, and alignment with business objectives.
Awareness of prompt injection risks and experience implementing mitigation strategies.
Familiarity with prompt tuning, parameter-efficient fine-tuning (PEFT), and prompt chaining methods.
Familiarity with continuous deployment and DevOps tools preferred. Experience with Git preferred
Experience working in agile/scrum environments
Successful track record interfacing and communicating effectively across cross-functional teams.
Good communication, analytical and presentation skills, problem-solving skills and learning attitude

Mandatory Key SkillsTensorflow,Pytorch,Java,Agile,Scrum,Devops,Ci/Cd,Jupyter Notebook,Aws Lambda,C++,Git,Aws Sagemaker,Devops Tools,Ml,Aws,Machine Learning,Deep Learning,Natural Language Processing,Artificial Intelligence,Neural Networks,Data Science,Keras,Python*