Skip to content
Saf9a

Case studies

Case studies in web development, DevOps, and AI automation

Case studies showing how Saf9a delivers web development, DevOps, and AI automation projects for product and operations teams.

Freight and supply chain

Atlas Logistics Visibility Portal

Built a real-time shipment visibility portal that reduced status requests and improved on-time reporting.

Problem

Operations relied on spreadsheets and manual updates, leading to delays, missed SLAs, and constant email churn.

Solution

  • Designed a unified tracking dashboard with live status updates.
  • Integrated carrier APIs and automated alert workflows.
  • Added role-based access and weekly reporting exports.

Results

  • 60% fewer status update emails
  • 3 hours saved per coordinator each day
  • Improved on-time reporting accuracy

Timeline

5 weeks

Tech stack

Next.js, Node.js, PostgreSQL, Docker, AWS

Impact

-60% support requests3x faster dispatch updates99.9% uptime
Screenshot placeholder
Dashboard placeholder

Multi-location retail

Zaytuna Retail Ops Suite

Delivered a retail operations suite with inventory forecasting, approvals, and performance dashboards.

Problem

Inventory was tracked in multiple tools with no single source of truth, causing stockouts and slow decisions.

Solution

  • Built an admin workspace with approvals and audit logs.
  • Introduced automated reorder triggers and weekly forecasts.
  • Connected POS data with live dashboards and alerts.

Results

  • 25% reduction in stockouts
  • 2x faster replenishment cycles
  • Leadership visibility across all stores

Timeline

4 weeks

Tech stack

Next.js, TypeScript, PostgreSQL, Vercel, GitHub Actions

Impact

25% fewer stockouts2x faster reportingZero downtime launches
Screenshot placeholder
Dashboard placeholder

Healthcare services

MedAssist Intake Automation

Automated patient intake with document processing and a RAG assistant for faster staff responses.

Problem

Staff handled intake forms manually, leading to slow turnarounds and inconsistent data quality.

Solution

  • Implemented document parsing with structured data validation.
  • Built a RAG assistant to answer policy questions instantly.
  • Automated task routing and follow-ups across the team.

Results

  • 70% reduction in manual data entry
  • Intake cycle cut from days to hours
  • Higher consistency in client records

Timeline

3 weeks

Tech stack

Next.js, FastAPI, Python, Vector DB, AWS

Impact

-70% manual work90% data accuracySame-day intake
Screenshot placeholder
Dashboard placeholder

Want to build something similar?

Tell us about your project and we will suggest the best approach.