Agentic AI Engineer - RAG Architecture - LLM Systems

Why Join AspenView?

At AspenView, we’re more than a nearshore IT partner—we’re a people-first, purpose-driven company that believes great culture drives great outcomes. We’re passionate about connecting talent and technology to deliver measurable value for clients—and meaningful career paths for our people.

Here’s what you can expect:

• Competitive base
• Comprehensive benefits and wellness support
• Flexible work model: hybrid, remote, or in-office
• Real growth opportunities and leadership visibility
• Inclusive, respectful culture that blends U.S. innovation with Colombian heart
• A company that listens, invests in you, and celebrates wins together

About the Role

In this role, you will lead the architecture and implementation of multi-step agentic workflows capable of reasoning, tool execution, and synthesizing complex clinical and operational data. You will engineer sophisticated RAG pipelines that bridge unstructured medical knowledge with enterprise EHR systems, establishing strict safety guardrails, hallucination controls, and HIPAA-compliant AI pipelines.

What You Will Do

• Agentic System Engineering: Architect and deploy autonomous agentic AI workflows and multi-agent systems (using frameworks like LangGraph, LlamaIndex, AutoGen, or CrewAI) capable of multi-step reasoning, dynamic tool usage, and clinical task execution.
• Advanced RAG Architecture: Design, optimize, and scale production RAG pipelines utilizing hybrid search (dense + sparse retrieval), re-ranking, query transformation, context compression, and semantic routing over multi-modal clinical and research datasets.
• LLM Fine-Tuning & Evaluation: Evaluate, fine-tune, and benchmark foundational models (e.g., Llama 3, Claude, GPT-4, Med-PaLM) for specialized healthcare tasks, establishing rigorous evaluation frameworks (e.g., Ragas, TruLens) for accuracy, groundness, and latency.
• Vector Database Infrastructure: Manage and optimize vector stores (Pinecone, Qdrant, Milvus, Weaviate, or pgvector) for high-concurrency, low-latency similarity searches and knowledge retrieval.
• Guardrails & HIPAA Compliance: Implement enterprise safety controls, input/output sanitization, and hallucination guardrails (e.g., NeMo Guardrails, Llama Guard) to ensure 100% HIPAA compliance and zero unauthorized PHI leakage.
• Cross-Functional Innovation: Collaborate closely with US-based clinical researchers, cloud architects, and product leads to translate complex healthcare needs into autonomous AI capabilities.

What You Bring

Experience

• 4+ years in software production engineering, with at least 2+ years dedicated to building LLM applications, agentic workflows, and RAG architectures in production.

Technical Expertise

• Agentic Frameworks & Orchestration: Strong mastery of modern AI orchestration ecosystems (LangChain, LangGraph, LlamaIndex, AutoGen, or CrewAI).
• Search & Retrieval Infrastructure: Deep knowledge of Vector Databases (Pinecone, Qdrant, Milvus, pgvector), embedding models, semantic chunking strategies, and re-ranking models (e.g., Cohere Rerank, BGE).
• Programming & Cloud AI Stack: Advanced proficiency in Python, with hands-on deployment experience on cloud platforms (AWS Bedrock / SageMaker or Azure OpenAI / Vertex AI).
• Model Evaluation & Guardrails: Experience implementing LLM observability, evaluation metrics (faithfulness, answer relevance), and safety frameworks.
• Healthcare Interoperability (Preferred): Familiarity with medical ontologies (SNOMED, LOINC, ICD-10) and healthcare data interfaces (FHIR APIs, HL7) is a strong plus.

Language Proficiency

• Advanced/Fluent English (C1/C2) for daily technical collaboration with Boston-based technology and research leadership.

Soft Skills & Competencies

• Systems Thinking for Non-Deterministic AI: Exceptional ability to debug, evaluate, and stabilize non-deterministic model outputs for mission-critical healthcare applications.
• Clinical & User Empathy: Deep commitment to designing transparent, explainable, and safe AI systems that empower clinicians rather than replace them.
• Proactive Nearshore Ownership: Self-starter capable of driving end-to-end AI features autonomously in a remote-first, agile nearshore setting.
• Clear Technical Storytelling: Ability to communicate complex AI concepts, architectural trade-offs, and evaluation metrics clearly to non-technical stakeholders.