resumeURL

Resume/CV

Name

Contact info

experienceYears

Work experience

noticePeriod

Notice period

holdingOffer

Holding offer?

"We serve high-compliance financial clients. Briefly explain how you would architect a RAG system to ensure Tenant Isolation (Client A cannot query Client B's vectors) and how you would handle PII Redaction before the data hits the LLM context window."

strata-core.ai

Founder @ Strata Core | INSEAD Alum | Building Sovereign AI for Global Supply Chains

The Mission: Sovereign CertaintyStrata Core is the autonomous intelligence layer for global supply chain compliance. In an era of "Compliance Drift," human-led audits are failing. We are building the Operating System for Regulatory Integrity—a multi-agentic workforce that ingests, audits, and monitors supply chain data with zero-trust precision.We are not building a chatbot. We are building a "Sovereign Data Fortress" that banks and governments will rely on. We need an Architect who understands that Accuracy is our currency and Privacy is our product.The Challenge (What You Will Build)You will report directly to the Founder (INSEAD Alum / Global Executive) and own the technical roadmap from Day 1.1. The Engine (Multi-Agent Orchestration)<ul><li>The Sensor: Build robust ingestion pipelines (OCR/Vision) for "messy" unstructured data (PDFs, stamps, handwritten logs) using tools like LlamaIndex or <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="http://Unstructured.io">Unstructured.io</a>.</li><li>The Auditor: Architect the logic layer where agents cross-reference data against complex rule sets (Halal, ESG, Tax) without hallucination.</li><li>The Sentinel: Design 24/7 monitoring agents that detect anomalies and trigger "Auto-Remediation" drafts (e.g., drafting emails to suppliers).</li></ul>2. The Fortress (Security &amp; Architecture)<ul><li>Sovereign Isolation: Implement Logical Tenant Isolation (Namespace Partitioning) in Vector Databases to ensure Client A’s data is invisible to Client B.</li><li>Privacy-First RAG: Design PII redaction pipelines (e.g., Microsoft Presidio) that scrub sensitive data before it touches the LLM context window.</li><li>Cost-Neutral Scaling: Build "Stateless" processing flows that minimize storage costs while maintaining audit trails.</li></ul>The StackWe are opinionated about using the best tools for the job:<ul><li>Core: Python, FastAPI/Django.</li><li>Orchestration: LangGraph, CrewAI, or AutoGen (Agentic Workflows).</li><li>Memory: Pinecone, Weaviate, or Qdrant (Vector Search).</li><li>LLMs: Azure OpenAI / AWS Bedrock (Enterprise Tier).</li><li>Frontend (MVP): Streamlit / V0 / Next.js.</li></ul>Who You Are<ul><li>The Builder: You have 5+ years of backend engineering experience and 2+ years exclusively in LLM/RAG production systems.</li><li>The Architect: You don't just write scripts; you think in systems. You know why "Physical Isolation" costs too much and how to secure "Logical Isolation" effectively.</li><li>The Operator: You thrive in an asynchronous, high-trust environment. You are comfortable sending a 5-minute Loom update at the end of your day rather than sitting in 3 hours of meetings.</li><li>The Pragmatist: You know when to use a regex script and when to burn tokens on GPT-4.</li></ul>The Offer<ul><li>Legacy: Build the infrastructure that protects the global food and ethical supply chain.</li><li>Ownership: Significant equity stake (vesting over 4 years).</li><li>Autonomy: You are the Technical Co-Founder in all but name. You define the stack; we provide the capital and the clients.</li></ul>How to ApplyTo ensure you are ready for this level of engineering, please answer the following question in your application note:"We serve high-compliance financial clients. Briefly explain how you would architect a RAG system to ensure Tenant Isolation (Client A cannot query Client B's vectors) and how you would handle PII Redaction before the data hits the LLM context window."

Founding AI Architect (Multi-Agent Systems)