AI Context Engineer

Role Overview

• Design and maintain context pipelines that retrieve, rank, compress, and assemble information for LLM inputs
• Define and enforce context schemas, prompt templates, and structured data contracts across AI features
• Develop strategies for dynamic context window management — balancing relevance, recency, and token budget
• Architect retrieval-augmented generation (RAG) systems, including chunking strategies, embedding selection, and reranking layers
• Craft, iterate on, and systematically test system prompts and few-shot examples that reliably elicit desired model behaviors
• Build evaluation frameworks to measure prompt quality, instruction-following accuracy, and output consistency
• Identify and mitigate failure modes such as hallucination, instruction drift, context confusion, and context stuffing
• Maintain a prompt library and versioning system to support reproducibility and regression testing
• Build internal tooling for context inspection, token counting, and prompt debugging
• Integrate with vector databases (e.g., Pinecone, Weaviate, pgvector) and semantic search infrastructure
• Implement memory systems, conversation state management, and session persistence for multi-turn AI applications
• Collaborate with DevOps/MLOps to deploy and monitor context pipelines in production environments
• Stay current with LLM research: context window advances, long-context models, sparse attention, and retrieval techniques
• Prototype and evaluate new approaches to context construction, summarization, and compression
• Document findings, failure analyses, and architectural decisions for the broader team
• Mentor junior engineers and contribute to internal knowledge sharing on prompt engineering

Requirements

• Secret Clearance or the ability to obtain a secret clearance
• Develop and maintain software requirements as it relates to user requirements and existing software defects
• Knowledge of DevSecOps, agile development and the context engineering lifecycle
• Knowledge and experience with JavaScript
• Knowledge of JSON
• Knowledge of source code versioning
• System Test and Evaluation, planning execution and management.
• 3+ years of professional software engineering experience, with 1+ years working directly with large language models in production.
• Demonstrable hands-on experience building or shipping LLM-powered applications or features.
• Experience with at least one major LLM API platform: OpenAI, Anthropic, Google Gemini, Cohere, or equivalent.
• Familiarity with prompt engineering techniques: chain-of-thought, few-shot prompting, structured output, tool/function calling.

Benefits

• healthcare
• wellness
• financial
• retirement
• family support
• continuing education
• time off benefits