P1C1TSTS• Act as the team champion for GitHub Copilot and Claude Code, integrating these tools into daily workflows to increase in developer productivity.Design and maintain "Context Environments" (indexing, .github/copilot-instructions.md, etc.) to ensure AI assistants produce code that strictly adheres to our architectural standards.Master complex prompt techniques-such as Chain-of-Thought for logic extraction and Few-Shot for architectural mapping-to guide LLMs through non-trivial coding tasksDevelop and deploy AI agents (using frameworks like LangChain or LlamaIndex) to automate the conversion of legacy code (e.g., Monolithic Java, COBOL) into modern microservices.Build and maintain autonomous "migration agents" that can reason through codebase dependencies, generate pull requests, and self-heal based on compiler feedback.Implement RAG (Retrieval-Augmented Generation) systems to "talk" to legacy documentation and extract business logic for re-implementation.Establish automated testing and linting pipelines to validate all AI-generated code, ensuring security and performance benchmarks are met before human review.Create reusable prompt templates and chunking strategies to ensure consistency across different application migration workstreams.Embed Generative AI workflows directly into the CI/CD pipeline to automate documentation updates and unit test generation.5+ years of experience in AI/ML development with focus on coding assistance tools & agentic systemsHands-on development experience using Github Copilot, Claude Code for rapid AI prototyping and implementationExperience with LangGraph and agentic workflow developmentKnowledge of LLM evaluation methodologies, testing frameworks, and quality assuranceUnderstanding of reasoning models, embedding models, and prompt optimization Experience with document processing and unit test generation Experience in any of the cloud platform (AWS / Azure / GCP)