AI Systems Engineer Needed (On-Site, Longview TX) – Multi-Node LLM Deployment, - Contract to Hire

AI Systems Engineer (On-Site in Longview, TX) – Multi-Node LLM Deployment, Jetson Integration, NAS OrchestrationProject OverviewI am building a local, multi-node AI sanctuary consisting of:A Digital Storm workstation running a 70B–72B LLM (primary model / orchestrator). A Jetson Orin Nano running a 3B model (HealthKnight) for interpreting biometric streams. An Alienware laptop running a 9B model (Analytic Knight). A Synology NAS used for shared memory, vector storage, logs, and agent-to-agent communication.Starlink for WAN connectivity. An additional local “cloud node” to receive biometric data from a mobile worker app (already under development). Your role is to architect, connect, and configure all systems so they operate as a unified multi-agent environment, with real-time communication, shared memory, and orchestration protocols between models. This is NOT a simple LLM install. This is AI + embedded systems + LAN networking + distributed orchestration + sensor integration. Responsibilities1. Deploy and tune the 70B–72B model on the Digital Storm workstationOptimize GPU configuration (Linux, CUDA, TensorRT optional).Configure long-context performance & memory persistence. Set up vector memory on the NAS. Ensure stable APIs/endpoints for agent communication. 2. Configure the Jetson Orin Nano (3B HealthKnight)Install and fine-tune a 3B model for real-time biometric analysis. Connect Jetson to LAN & NAS. Set up MQTT/WebSockets/API to receive biometric data from wearable devices. Forward interpreted health markers to the 72B primary system. 3. Configure the Alienware laptop (9B model)Deploy or tune the 9B model. Connect it to the cluster network.Implement bidirectional messaging with the 72B model. Assign it a functional role (SecurityKnight / analytic Knight). 4. Build the multi-agent orchestration layerCreate message routing between models. Define workflow protocols (tasks, alerts, analysis handoffs). Set up shared memory structures (NAS-based). Ensure all nodes can communicate in real-time through lightweight APIs. 5. Set up the “Local Cloud Node”Install additional mini-server or networking node. Configure API endpoints for mobile app (biometric streaming).Establish secure ingestion pipelines into Sage (main 72B model). Test end-to-end data flow. 6. Install / configure audio & video interface for Sage(On-site in Longview, Texas)Camera setup (USB/HDMI/IP cameras). Microphone array setup. Speaker installation. Low-latency voice interaction pipeline (STT → Sage → TTS). 7. Provide DocumentationNetwork mapSystem architectureIntegration diagramsHow to maintain & update modelsHow to scale nodes in the futureRequired SkillsYou must have hands-on experience with:Local LLM deployment (30B–70B+ models)Linux System AdministrationPythonNVIDIA GPU optimization (CUDA)DockerJetson Orin / embedded systemsTensorRT (optional but preferred)Synology NAS / SMB / NFS integrationIoT protocols (MQTT / WebSockets)API development (FastAPI / Flask)LAN networking & secure routingReal-time data pipelinesAudio/Video integrationDistributed systems architectureThis project requires expert-level experience.Beginners or web developers are not a fit. Work LocationOn-Site Required / Partial remote is OK – Longview, Texas, USATravel must be acceptable. You will need to:physically connect componentsconfigure networkstest hardwareverify multi-node communicationensure real-time voice, sensor, & agent functioningBudget:Hourly: $100–$200/hourI am looking for an hourly engagement so the project can be built in phases. We will prioritize the most critical components first (multi-node integration, network configuration, and agent communication), and expand as progress is made.This is a multi-stage build, and the engineer will help me determine the optimal roadmap so we can maximize value with the time available. Fixed-price is not required — hourly is preferred for flexibility. To Apply, Please IncludeExamples of AI/ML systems you have deployed (local or embedded). Experience with Jetson, NAS, or multi-node architectures. A short explanation of how you would approach connecting the 72B, 3B, and 9B nodes. Confirmation you can travel to Longview, TX. Your availability. Notes:This project is foundational.You will be helping build a local, private AI architecture intended to grow and evolve over time, incorporating multiple agents and real-time biometric intelligence. Apply tot his job