Computer Vision MLE

About the Role

Mercor is seeking a senior Computer Vision Machine Learning Engineer (MLE) for a remote, contingent W2 engagement focused on evaluating the feasibility of a computer vision system that identifies and grades physical objects from images.

The initial engagement is expected to last 3–4 weeks and centers on performing a technical assessment, benchmarking model performance, evaluating data quality, determining realistic accuracy expectations, and producing executive-level recommendations. Successful performance may lead to a longer-term implementation engagement.

Key Responsibilities

Computer Vision Feasibility Assessment

• Evaluate the viability of a computer vision solution for identifying and grading physical objects from images
• Analyze business requirements and technical constraints
• Determine realistic system capabilities and limitations
• Assess production readiness and deployment feasibility

Model Evaluation & Benchmarking

• Benchmark baseline model performance using representative image datasets
• Fine-tune and evaluate modern vision foundation models
• Compare multiple model approaches and architectures
• Establish performance baselines and success criteria

Dataset Analysis

• Review image dataset quality and completeness
• Assess labeling accuracy and consistency
• Identify data gaps, biases, and quality concerns
• Evaluate dataset suitability for production deployment

Accuracy Measurement

• Build and maintain train/evaluation data separation
• Measure model performance against held-out validation sets
• Calculate accuracy, precision, recall, and other relevant metrics
• Perform calibration and reliability assessments

Performance Ceiling Analysis

• Determine realistic upper-bound performance expectations
• Identify factors limiting model accuracy
• Estimate potential improvements through additional data or modeling techniques
• Evaluate scalability and long-term performance potential

Technical Due Diligence

• Assess overall system feasibility
• Identify implementation risks and constraints
• Evaluate infrastructure and deployment requirements
• Recommend go/no-go decisions based on evidence

Executive Reporting

• Translate technical findings into business-focused recommendations
• Create decision-grade reports for non-technical stakeholders
• Present key risks, opportunities, and feasibility conclusions
• Communicate model performance and limitations clearly

Required Qualifications

• 5+ years of experience in Computer Vision, Machine Learning Engineering, or Applied AI Systems
• Hands-on experience fine-tuning vision foundation models and modern computer vision architectures
• Proven experience with image classification, object identification, quality grading, and defect detection
• Strong evaluation methodology including representative sampling, train/test separation, benchmarking, calibration, and performance validation
• Ability to assess technical feasibility, production readiness, and operational risks
• Strong written and verbal communication skills for executive audiences

Preferred Qualifications

• Authentication systems, counterfeit detection, anomaly detection
• Private equity technical due diligence or advisory engagements
• Imaging hardware systems, camera systems, lighting pipelines
• Edge or on-premise ML deployment
• Production ML infrastructure

Engagement Details

• Employment: W2 Contract (contingent engagement)
• Location: Remote (global)
• Initial Duration: 3–4 weeks
• Extension Potential: Yes — strong possibility of a longer-term implementation engagement
• Pay: $135/hr

About Mercor

Mercor partners with leading AI labs and enterprises to train frontier models and deliver advanced AI expertise. Contributors work on high-impact projects involving machine learning, evaluation, benchmarking, and AI system development.