Skip to content
View stmontgo123's full-sized avatar

Achievements

Achievement: YOLO

Achievements

Achievement: YOLO

Block or report stmontgo123

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
stmontgo123/README.md

Featured Project

NVIDIA AI Runtime Reliability Lab

A hands-on enterprise AI inference reliability lab comparing PyTorch, TensorFlow/Keras, and ONNX Runtime, with a production path toward NVIDIA NIM, Triton Inference Server, TensorRT, GPU telemetry, Kubernetes/Run:ai, and enterprise observability.

Focus areas: NVIDIA AI infrastructure · AI inference · Runtime reliability · Benchmarking · SRE · Python · Infrastructure automation · Observability

The lab includes:

  • PyTorch, TensorFlow/Keras, and ONNX Runtime inference workflows
  • Repeatable latency and throughput benchmarking
  • Prediction consistency checks across runtimes
  • Simulated traffic-spike and degraded-node scenarios
  • Model-version and rollback simulation
  • Synthetic infrastructure-failure prediction data
  • Streamlit executive dashboard
  • Architecture, runbook, decision-defense, executive talk track, LinkedIn, and presentation documentation
  • A production architecture path toward NVIDIA NIM, Triton, TensorRT, GPU telemetry, Kubernetes/Run:ai, and Prometheus/Grafana

Design principle: The best AI model is not automatically the best production AI service. Runtime choice, portability, observability, recovery behavior, and infrastructure architecture determine whether inference can meet enterprise reliability objectives.

Repository: NVIDIA AI Runtime Reliability Lab

Pinned Loading

  1. secure-agentic-ai-banking-demo secure-agentic-ai-banking-demo Public

    Secure agentic AI banking proof of concept using Oracle AI Database 26ai, vector search, RAG, Python, Ollama, role-aware retrieval, human approval, and auditability.

    Python

  2. careguard-ai-healthcare careguard-ai-healthcare Public

    Secure HIPAA-aligned RAG and agentic AI healthcare demo using Oracle Database 26ai, vector search, local LLM, human oversight, and auditability.

    Python

  3. nvidia-opsguard-ai nvidia-opsguard-ai Public

    Governed agentic AI for secure enterprise IT operations using NVIDIA Nemotron, RAG, human approvals and audit logging.

    Python

  4. nvidia-ai-factory-guardian nvidia-ai-factory-guardian Public

    Agentic SRE + FinOps simulator for NVIDIA AI factories—detecting failures, rebalancing GPU workloads, improving token economics, and preserving human approval with an auditable trail.

    Python

  5. stop-hallucinating-rag stop-hallucinating-rag Public

    Executive-friendly RAG demo showing LLM hallucination vs grounded answers, plus bad RAG data vs good RAG data.

    Python

  6. nvidia-ai-runtime-reliability-lab nvidia-ai-runtime-reliability-lab Public

    Enterprise AI inference reliability lab comparing PyTorch, TensorFlow, and ONNX Runtime, with a production path to NVIDIA NIM, Triton, TensorRT, and GPU-scale AI infrastructure.

    Python