The Vision
We aren’t looking for a “Data Scientist” to build dashboards. We need an architect who understands how to turn unstructured global data into a “Sovereign Knowledge Graph” that autonomous agents can query with high reasoning accuracy.
Strategic Function in the Tiger Team
This person builds the
“Shared Memory” of the swarm. Without them, your agents are just fast-talking chatbots. With them, the agents have a persistent, sovereign intelligence they can all tap into.
Job Description: Lead Reasoning & Knowledge Architect
The Role: As a founding member of the OPTe Tiger Team, you will architect the “Sovereign Knowledge Layer.” You are responsible for ensuring our autonomous agentic swarms don’t just “hallucinate,” but reason against a grounded, high-fidelity knowledge graph. You will bridge the gap between Large Language Models and structured symbolic logic.
What You’ll Build:
- Shared Sovereign Memory: Design a distributed vector and graph-based memory system that allows agents to share context across the “Follow-the-Sun” nodes.
- Reasoning Chains: Implement and optimize advanced reasoning frameworks (Chain-of-Thought, Tree-of-Thought) to ensure agents can handle complex, multi-step missions.
- Small Model Optimization: Fine-tune and deploy efficient, specialized models (Mistral, Llama-3) that can reason at the edge without the latency of massive API-based models.
Who You Are:
- The “Graph” Expert: Deep experience with Qdrant, Weaviate, or Neo4j. You know how to make vectors and graphs talk to each other.
- Systems Mindset: You treat AI as a distributed systems problem. You understand the latency challenges of RAG (Retrieval-Augmented Generation) at scale.
- European Pedigree: Ideally coming out of a high-complexity environment like CERN, DeepL, Mistral, or a Tier 1 Quant firm.
The Conditions: This is a high-autonomy, founding-team role. We move fast, we pay top-tier, and we operate in a “First Who, Then What” culture. No bureaucracy—just pure engineering for the sovereign intelligence layer.
