Senior Machine Learning Engineer

As a Senior Machine Learning Engineer with expertise in Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG), you will play a critical role in supporting a high-impact Proof of Concept (PoC) focused on legacy code analysis, transformation, and modernization. Your primary responsibility will be to enable intelligent code migration and documentation through the implementation of advanced AI/ML tooling. You will be expected to develop and integrate LLM-based pipelines, utilizing tools such as Claude Sonnet 3.7 or 4 on AWS Bedrock. Additionally, you will design and implement RAG-based systems for code understanding, leveraging Vector Databases like Milvus or Pinecone. Your expertise in Abstract Syntax Tree (AST) techniques will be crucial for parsing, analyzing, and transforming legacy code for migration and documentation purposes. You will also apply CodeRAG techniques to facilitate context-aware retrieval and transformation of code artifacts. Iterative validation and correction of AI-generated outputs will be part of your routine to ensure the production of high-quality code assets. Data preprocessing and metadata enrichment, including embeddings, structured knowledge, or fine-tuning for LLM input optimization, will also be within your scope of work. Collaboration with domain experts and engineering teams is essential to ensure alignment with architecture and business logic. You will utilize version control systems like Git to manage changes, support collaboration, and ensure reproducibility. Additionally, you will contribute to QA strategies and help define testing protocols for model output validation. To excel in this role, you must have at least 5 years of experience in Machine Learning, with a strong focus on LLM applications and code understanding. Proficiency in Python and solid software engineering principles are a must, as well as experience working with AWS Bedrock for model deployment and orchestration. You should possess a strong understanding and hands-on experience with AST-based code parsing and transformation, familiarity with RAG architectures, and experience working with vector databases like Milvus, Pinecone, or similar. Experience in preprocessing legacy codebases, enriching metadata for LLM consumption, and using Git or other version control systems in collaborative environments are crucial skills. A solid understanding of code migration, modernization processes, and business logic documentation is also required. Nice-to-have skills include ensuring compliance with architectural and code specifications, documenting code flows, aligning with business requirements, familiarity with QA and testing strategies in AI/ML or code-generation workflows, and a collaborative mindset with strong communication skills and a proactive attitude essential for working in a fast-paced PoC environment with tight feedback loops.,