Data Analyst – Portcast

Portcast Data Analyst Role | Delhi / Remote | Python, SQL & Predictive Analytics

Opportunity Overview

Data Quality & Predictive Analytics Workflow

The Data Analyst at Portcast owns end-to-end data quality for real-time ocean freight predictions (ETA/demurrage risk), investigating data anomalies, building scalable QA workflows, and feeding clean features back into ML models:

┌───────────────────────────────────────────┐
│ 1. Multi-Source Ingestion & Monitoring    │ ➔ Ingest & audit ocean container tracking feeds across global shipping lines
└─────────────────────┬─────────────────────┘
                      ▼
┌───────────────────────────────────────────┐
│ 2. Root-Cause Anomaly Investigation       │ ➔ Detect data gaps, trace prediction errors to source, & engineer systemic fixes
└─────────────────────┬─────────────────────┘
                      ▼
┌───────────────────────────────────────────┐
│ 3. Automated Python/SQL Pipeline QA       │ ➔ Build reusable scripts to automate prediction accuracy checks and report generation
└─────────────────────┬─────────────────────┘
                      ▼
┌───────────────────────────────────────────┐
│ 4. Product, ML & Customer Enablement      │ ➔ Supply clean feature sets to Data Science and deliver accuracy dashboards to commercial teams
└───────────────────────────────────────────┘

Key Responsibilities

• Data Quality & Anomaly Ownership: Audit ocean freight data feeds to spot prediction inaccuracies (delay predictions, detention/demurrage risks) and perform root-cause analysis before clients flag issues.

• Reusable Automation Workflows: Write modular, scalable Python code and SQL queries to automate internal accuracy metrics, pipeline checks, and visualization dashboards.

• Data Science & Feature Engineering Support: Partner with ML Engineers and Data Scientists to optimize prediction model inputs by fixing underlying data structures and introducing new operational features.

• Cross-Functional Analytics Delivery: Serve as the analytics bridge for Customer Operations and Product teams, maintaining customer-facing dashboards covering prediction explainability, coverage, and timeliness.

• AI-Assisted Analytics Execution: Leverage AI coding assistants (Claude, Copilot) responsibly to accelerate QA, script generation, and documentation while independently verifying data outputs.

Qualification Matrix & Technical Skill Stack

Core Requirements

Key Focus Areas for Interview Preparation

1. Root-Cause Anomaly Detection: Prepare case studies demonstrating how you discovered a data pipeline error, traced it back to raw source ingestion, and built an automated fix to prevent recurrence.

2. First-Principles System Design: Be ready to discuss trade-offs in your past data stack decisions—explaining why you chose a specific SQL or Python design pattern over introducing external tools.

3. Logistics & Supply Chain Metrics: Review core logistics data metrics such as ETA Variance, Container Dwell Time, Demurrage & Detention Costs, Vessel Transshipment Delays, and Prediction Accuracy Scorecards.