Stutz A.

Key Skills

Software

Remotasks

Mindrift

Mercor

Data Annotation Tech

Top Subject Matter

Software Engineer

Data Analysis

E-Commerce

Top Data Types

Image

Text

Video

Top Task Types

Segmentation

Classification

Question Answering

Freelancer Overview

I am a Senior Software Engineer with over four years of production experience across cloud architecture, distributed systems, and machine learning integration, now directing that technical depth toward AI training and model evaluation work. My background gives me an unusually grounded perspective on what makes AI systems behave reliably in production, having spent years building the data pipelines, API contracts, and infrastructure layers that frontier models ultimately depend on. At Salesforce I worked directly alongside machine learning engineers to bring AI-powered systems into production at scale, which gave me firsthand understanding of how model outputs need to be structured, evaluated, and refined through human feedback to perform consistently in real-world conditions. What I bring to AI training specifically is the combination of strong analytical reasoning, rigorous attention to technical detail, and the ability to evaluate complex outputs against precise criteria without losing sight of the broader context those outputs need to serve. My academic foundation in computer science from Carnegie Mellon, combined with deep practical experience in Python, distributed data systems, and system design, means I can engage meaningfully with technically demanding tasks across software engineering, data analysis, and reasoning-heavy evaluation projects. I approach every task with the same standard I apply to production engineering work, which is that quality, consistency, and intellectual honesty are not optional extras but the baseline from which everything else is built.

Labeling Experience

Data Annotation

ImageClassification

This project involved image classification and object detection labeling for computer vision model training at Data Annotation, covering a dataset of over 120,000 images across multiple annotation cycles spanning healthcare diagnostics, autonomous systems, and consumer product recognition. Core tasks included drawing precise bounding boxes, polygon segmentation masks, semantic category tagging, and multi-attribute labeling across diverse image classes. Quality was maintained through a structured review pipeline where each labeled batch underwent inter-annotator agreement checks before submission, consistently achieving agreement scores above 92%. All work was completed in compliance with the platform's tiered quality assurance framework, with escalation protocols followed for edge cases involving occlusion, ambiguous class boundaries, and low-resolution inputs.

2022 - 2026

Education

Work History