Tech Stack

Description

You will define the long-term technical architecture and strategy for SoFi's next-generation AI platform, focusing on scalable agentic frameworks and LLM deployment. You'll architect graph-based LLM orchestration with LangGraph, build persistent agent state infrastructure, and pioneer advanced model optimization techniques. This role directly impacts risk management and compliance, helping millions of members achieve financial success.

Requirements

• Bachelor's or Master's in CS, Data Science, AI, ML, or related field (PhD a plus)
• 8+ years software development, 3+ years deploying production AI applications
• Expert-level experience with LangGraph for complex multi-step LLM orchestration
• Expert-level proficiency in developing sophisticated agentic solutions with planning, memory, and tool integration
• Deep expertise in LLM architectures, prompt engineering, RAG, and PEFT techniques (e.g., LoRA)

Responsibilities

• Define long-term technical architecture and strategy for AI platform including agentic frameworks and LLM deployment patterns
• Architect and standardize graph-based LLM orchestration using LangGraph for complex multi-stage reasoning
• Develop persistent infrastructure for distributed agent memory and state management
• Pioneer and institutionalize parameter-efficient fine-tuning (PEFT) and model compression techniques
• Build unified model serving platform ensuring high-throughput, low-latency inference