Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/ML engineer specializing in end-to-end system delivery across training, explainability, and production deployment. I build real-time computer vision and predictive ML pipelines—from feature engineering and model evaluation to REST APIs, monitoring dashboards, and containerized/cloud deployments. I’m comfortable owning projects end-to-end with a focus on shipping production-grade AI solutions. My work spans FastAPI, PyTorch, Docker/Kubernetes, SQL data layers, and modern CV/LLM tooling (YOLOv8, CUDA acceleration, Hugging Face) with explainability (SHAP) and performance-minded inference.…AI/ML engineer specializing in end-to-end system delivery across training, explainability, and production deployment. I build real-time computer vision and predictive ML pipelines—from feature engineering and model evaluation to REST APIs, monitoring dashboards, and containerized/cloud deployments. I’m comfortable owning projects end-to-end with a focus on shipping production-grade AI solutions. My work spans FastAPI, PyTorch, Docker/Kubernetes, SQL data layers, and modern CV/LLM tooling (YOLOv8, CUDA acceleration, Hugging Face) with explainability (SHAP) and performance-minded inference.

NEURALFORGE

AI Engineer, Software Engineer, AI Developer, +3





AI/ML engineer specializing in end-to-end system delivery across training, explainability, and production deployment. I build real-time computer vision and predictive ML pipelines—from feature engineering and model evaluation to REST APIs, monitoring dashboards, and containerized/cloud deployments. I’m comfortable owning projects end-to-end with a focus on shipping production-grade AI solutions. My work spans FastAPI, PyTorch, Docker/Kubernetes, SQL data layers, and modern CV/LLM tooling (YOLOv8, CUDA acceleration, Hugging Face) with explainability (SHAP) and performance-minded inference.…AI/ML engineer specializing in end-to-end system delivery across training, explainability, and production deployment. I build real-time computer vision and predictive ML pipelines—from feature engineering and model evaluation to REST APIs, monitoring dashboards, and containerized/cloud deployments. I’m comfortable owning projects end-to-end with a focus on shipping production-grade AI solutions. My work spans FastAPI, PyTorch, Docker/Kubernetes, SQL data layers, and modern CV/LLM tooling (YOLOv8, CUDA acceleration, Hugging Face) with explainability (SHAP) and performance-minded inference.

Available to hire

AI/ML engineer specializing in end-to-end system delivery across training, explainability, and production deployment. I build real-time computer vision and predictive ML pipelines—from feature engineering and model evaluation to REST APIs, monitoring dashboards, and containerized/cloud deployments.

I’m comfortable owning projects end-to-end with a focus on shipping production-grade AI solutions. My work spans FastAPI, PyTorch, Docker/Kubernetes, SQL data layers, and modern CV/LLM tooling (YOLOv8, CUDA acceleration, Hugging Face) with explainability (SHAP) and performance-minded inference.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Work Experience

AI/ML Engineer (Contract) at Enterprise Software Client

January 1, 2026 - December 31, 2026

Engineered an AI-powered field-sales route optimizer and sales forecasting system. Built a priority-weighted route recommendation workflow using live traffic data combined with an XGBoost buying-probability model trained on 15 engineered features. Achieved 0.91 ROC-AUC and used SHAP for explainability, identifying historical conversion rate as the dominant driver (43% importance). Implemented forecasting with Prophet + XGBoost behind a RESTful FastAPI API and delivered dashboards for forecast visualization and model-health monitoring. Integrated with a cloud SQL backend to support near real-time customer data updates.

AI/ML Research Engineer at Collaborative University-Industry Research Program

January 1, 2025 - December 31, 2025

Designed and built a real-time Helmet & License Plate Recognition pipeline for automated traffic-violation detection. Developed and fine-tuned models using YOLOv8 for detection and PaddleOCR for OCR. Reached 82% mAP50 for helmet detection and 88% accuracy for license-plate recognition. Optimized inference performance using PyTorch with CUDA acceleration for efficient real-time video-stream processing.