Operations

_________________

Data Annotation & Engineering

Purpose: To accurately and efficiently annotate large sets of data, including images, videos, etc. We secure specialized resources when needed.

Client Intake

• Project Review

• Client Demo

• Company Demo

• Sign-off

Pre-Annotation

• Team Sourcing

• Guideline Training

• Training Queue

Annotation & Review

• Annotation

• QA Review

• Daily Standup

• Weekly Calibration

Post-Annotation

• Internal Report

• Client Report and Feedback

Client Intake (with Client Engineers/Quality Assurance)

• Email

• Meeting

  • Scope of Project - What’s the application?

  • Training/Production of Data Set (API Key/File)

  • Tooling - internal, etc

  • Average Handling Time

  • QA Process/Metrics

  • Feedback Loop and Channel

  • Completion Date

Data Collection & Ingestion

• Collection the data

Annotation

• Team Alignment (Look at Guidelines Together, training on tool) 

• Daily Sync (30 min) 

• Annotation Production 

• Check-in (Team Lead) 

• QA Audit` 

• Feedback (quality metrics) (Team Lead and QA) 

• Escalation (to clients) whole project 

• Recalibration Meetings (Team Lead) 

• Weekly Business Meetings (Report to Client from Team Lead) 

Final Delivery

• Format to the ML (Recommended to the client) YOLO 8, 9, 11

Post-Project Report

• Internal Analysis

• External Analysis (Client Review)