Junior AI Engineer, Full Stack Developer

• Add AI capabilities, LLM-powered features, workflows, and interfaces to existing full-stack applications.
• Build and improve retrieval-augmented generation (RAG) systems with a Senior AI Engineer.
• Develop chunking and embedding strategies and evaluate retrieval quality.
• Manage context windows and model inputs across text, documents, and images.
• Help design agentic workflows involving multi-step LLM pipelines, tool use, and orchestration.
• Prompt-engineer and evaluate LLM workflows.
• Write clean, testable services and data pipelines moving information between systems.
• Translate workflow needs from investment and operations colleagues into concrete technical problems.
• Take sprint tasks directly, contribute to active RAG or agentic workflow projects, and increasingly own workflow components.
• Contribute hands-on to evaluation and testing pipelines.
• Work on a small embedded technology team and apprentice closely with a Senior AI Engineer.

• Bachelor’s or Master’s in Computer Science, AI/ML, Engineering, or a related field.
• Solid software engineering fundamentals in Node.js and Type Script, including REST APIs, async and promise-based concurrency, a testing framework such as Jest or Vitest, and everyday comfort in git.
• Familiarity with AWS Amplify is a plus.
• Python is a nice to have, especially for RAG and data-heavy work.
• Exposure to cloud infrastructure, ideally AWS, including Bedrock, Bedrock Agent Core, and Knowledge Bases for Bedrock.
• Practical experience with LLM APIs such as OpenAI or Anthropic, including function calling and tool use, structured outputs, streaming responses, and token/context-window management.
• Hands-on experience building a real LLM project, prototype, or shipped feature.
• Experience designing LLM-in-the-loop evaluation and testing pipelines, including golden datasets, LLM-as-judge scoring, regression tests, and tools such as promptfoo, Braintrust, or Ragas.
• Comfort with AI-native development tools such as Claude Code or Cursor.
• Ability to measure prompt, retrieval, and model changes using evaluation scores, latency, or cost-per-call metrics.
• Coachability and initiative.
• Clear communication with non-technical investment and operations professionals.
• Must be authorized to work in the U.S. without current or future employer sponsorship; employment visa sponsorship is unavailable, including sponsorship after F-1 OPT/CPT.
• Bonus: familiarity with RAG mechanics, vector stores, agent/RAG frameworks, production AI features, data sensitivity, and the Model Context Protocol (MCP).

Demonstrates expertise in developing and integrating AI capabilities into full-stack applications, with a strong focus on LLM workflows, retrieval-augmented generation systems, and effective communication with cross-functional teams. Proficient in software engineering principles, particularly in Node.js and Type Script, while leveraging cloud infrastructure and LLM APIs for innovative solutions.

• Node.js Development
• Type Script Programming
• LLM API Experience
• Retrieval-Augmented Generation (RAG)
• Cloud Infrastructure (AWS)

• Software Engineering Fundamentals
• REST APIs
• Async and Promise-Based Concurrency
• Testing Frameworks (Jest, Vitest)
• Data Pipelines
• Chunking and Embedding Strategies
• Evaluation and Testing Pipelines
• Prompt Engineering
• Model Context Management
• AI-Native Development Tools

• Clear Communication
• Coachability
• Initiative

• AI/ML
• Investment Operations
• Data Sensitivity
• Model Context Protocol (MCP)
• Agent/RAG Frameworks

• AWS Amplify
• OpenAI
• Anthropic
• Claude Code
• Cursor
• Promptfoo
• Braintrust
• Ragas