ML / MLOps Engineer • Telemetry & Reliability
Building GPU training reliability, telemetry pipelines, and secure ML in production-like environments.

LinkedIn • Phoenix, Arizona · Open to work

• Design and ship end‑to‑end ML systems: data ingest, modeling, evaluation, and deployment with Docker and modern Python stacks.
• Build telemetry and anomaly detection pipelines for aerospace, ESG, and industrial‑style workloads.
• Focus on reliability and safety: GPU OOM mitigation, health monitoring, and audit‑friendly analytics.

Predictive maintenance pipeline using NASA C‑MAPSS engine data for RUL forecasting and anomaly detection, with Dockerized local deployment.

• Implements time‑series modeling and anomaly detection for turbofan degradation signals.
• Ships with an MLOps‑oriented structure: clear data, models, and experiments layout plus containerized runtime.
• Good fit for roles touching reliability engineering, telemetry ML, or industrial analytics.

🔗 Repo: PulseNet

Mission‑control‑style telemetry simulation and anomaly surfacing stack for orbital monitoring and operator dashboards.

• Encodes satellite‑like telemetry streams with anomaly flags for critical subsystems.
• Provides Streamlit dashboards so operators can inspect timelines, health states, and events.
• Demonstrates how to turn raw telemetry into actionable situational awareness.

🔗 Repo: CommandX

CubeSat telemetry monitoring pipeline for anomaly detection, Firebase‑backed data flow, and satellite health analytics.

• Ingests and structures CubeSat‑style telemetry and runs anomaly detection over subsystems.
• Uses Firebase‑backed flows to simulate cloud‑native satellite operations.
• Pattern generalizes to any sensor‑heavy robotic or edge system.

🔗 Repo: orbit-Q

Prototype for modeling GPU memory fragmentation and evaluating strategies to reduce training‑time out‑of‑memory failures on RTX‑class workloads.

• Models fragmentation patterns under deep learning training workloads.
• Evaluates mitigation strategies for OOM reduction and better GPU utilization.
• Relevant for ML platform, infra, and reliability engineering teams.

🔗 Repo: Predictive-GPU-Memory-Defragmenter

ESG telemetry and carbon analytics platform for async pipelines, forecasting, anomaly detection, and audit‑friendly reporting.

• Tracks portfolio‑level ESG metrics, carbon trends, and anomalies.
• Demonstrates async data pipelines, forecasting, and Postgres‑backed analytics.

🔗 Repo: Eco-Enterprise

• ML / MLOps roles with a focus on telemetry, observability, and reliability.
• Applied ML in aerospace, robotics, ESG, and safety‑critical systems.
• Teams that value clean experiments, clear metrics, and production‑minded tooling as much as model accuracy.

• Languages: Python
• ML: PyTorch, time‑series modeling, anomaly detection
• MLOps / Systems: Docker, Streamlit, telemetry pipelines, basic cloud primitives (e.g., Firebase, AWS)
• Domains: Telemetry, aerospace‑style monitoring, GPU reliability, ESG analytics