Member of Technical Staff

Responsibilities

• Design and build agentic systems that plan, delegate, reflect, and coordinate across tools, models, and environments
• Develop memory and context management systems including long-term, episodic, and working memory with retrieval-augmented reasoning
• Create multimodal intelligence systems using VLMs, VLAs, and image/video retrievers for visual and embodied environments
• Build efficient inference infrastructure for agentic workloads including KV caching, speculative decoding, and model routing
• Prototype new ideas quickly and advance promising work into production-grade systems
• Define evaluation methods for agentic behavior, memory coherence, multimodal grounding, and inference reliability

Requirements

• Track record building serious systems with LLMs, agents, retrieval, multimodal models, or inference infrastructure
• Comfort reading research papers, writing code, running experiments, and debugging complex systems
• Ability to operate in ambiguity and define problems without perfect problem statements
• Strong balance of conceptual novelty and engineering taste
• Courage to explore unconventional ideas and discipline to abandon weak ones

Preferred Qualifications

• Background in LLM agents, planning, tool use, multi-agent systems, or agentic orchestration
• Experience with RAG, memory systems, context engineering, or retrieval systems
• Work with VLMs, VLAs, robotics, video understanding, or embodied AI
• Expertise in model evaluation, verifier design, reward modeling, or benchmark construction
• Experience in inference optimization, serving systems, distributed systems, or search infrastructure
• Reinforcement learning, applied research engineering, or 0-to-1 system building experience

About the Company

Superintelligence Lab is building the next generation of intelligent systems: agents that can reason, remember, search, use tools, interact with multimodal worlds, and execute complex work with reliability. The team values taste over trend-following, depth over noise, speed with judgment, and high agency with low ego.