AI Architect

·         Agentic System Architecture: Define the technical standards, reference designs, and operating models for multi-agent systems, including Agentic Workflows, Task-Based Evals, and Memory Management.

·         RAG and Knowledge Integration: Architect Retrieval Augmented Generation (RAG) pipelines to connect LLMs with proprietary scientific data, knowledge graphs (like Neo4j), and vector databases (Pinecone).

·         Tool-Use & API Integration: Build frameworks allowing agents to interact securely with external tools, laboratory equipment (IoT), LIMS (Laboratory Information Management Systems), and regulatory databases.

·         Platform Development: Implement "AI Gateway" architectures to ensure model agnosticism (using multiple LLMs) and build scalable, cloud-native infrastructures (Azure/AWS).

·         Safety and Compliance: Establish guardrails to prevent data leakage, prompt injection, and hallucination in regulated environments (FDA, HIPAA, GDPR).

Core Competencies and Skills

• AI Frameworks: Deep experience with Agentic frameworks such as LangChain, LangGraph, AutoGen, CrewAI, or Microsoft Semantic Kernel.

1. Programming: High proficiency in Python and experience with C#.

• LLMs & NLP: Expertise in deploying LLMs (GPT-4, Claude, Llama 3) via APIs and fine-tuning models for domain-specific tasks.

1. Cloud & MLOps: Experience with AWS/Azure, Kubernetes, Docker, and CI/CD pipelines for AI systems.