Darren Sheffield

Selected project work

Systems work across recovery, rollout, integration, and automation.

These examples show how I approach operational systems: isolating the real failure mode, turning ambiguous symptoms into evidence-backed plans, and leaving teams with clearer workflows, reporting, documentation, and ownership.

Recovery and stabilization

Post-go-live systems under pressure

Projects where the main job was to turn ambiguous symptoms into a concrete, supportable recovery path.

ERP customer portfolio

NetSuite performance recovery and executive-ready remediation plan

A customer-facing recovery effort for severe NetSuite performance problems after active operational use. The work required separating user pain, configuration patterns, reporting load, and customization behavior into findings that business and technical stakeholders could act on.

NetSuite APM SuiteAnalytics Saved searches Benchmark reports Stakeholder briefings

Work performed

Built benchmark packages, reviewed transaction and page-load patterns, translated technical evidence into prioritized remediation, and kept the discussion tied to customer value and renewal health.

Outcome

Helped reduce severe page-load experiences from roughly 30-40 seconds toward the 10-15 second range while giving stakeholders a clearer path for continued optimization.

Commerce integration portfolio

iPaaS error remediation across order, shipment, and return flows

Stabilized recurring integration failures across marketplace, ecommerce, EDI, fulfillment, and finance-facing reconciliation paths where each error needed both technical diagnosis and business-impact translation.

  • Resolved 30+ integration error scenarios across Amazon, Shopify, EDI, and high-volume order flows.
  • Used payload evidence, mapping review, retry behavior, and ownership clarity to reduce repeat failures.
  • Improved confidence in order, shipment, and finance/audit reconciliation workflows.
Celigo integrator.io NetSuite Shopify EDI Payload analysis

Integration operations prototype

Local monitor for NetSuite to Shopify failure handling

Built a local-first reference implementation for tracking integration errors, mapping rows, payload snapshots, retry attempts, transform hooks, and resolution state across NetSuite to Shopify flows.

Blurred interface capture from a local integration monitor prototype
Prototype interface example for integration triage and retry workflows.
  • Modeled failure cases in a browser-runnable tool that can be opened without a hosted backend.
  • Preserved editable mapping and transform history so fixes could be reviewed instead of guessed.
  • Created a clearer workflow for triage, retry, and resolution state during integration support.
Static HTML JavaScript localStorage REST payloads Retry state

Rollout and operating model

Making systems usable after launch

Projects where success depended on documentation, ownership, training, controls, and maintainable handoffs.

Consumer operations business

ERP deployment and business-systems operating model

Owned systems administration and change management for a distributed operations environment, including a full NetSuite deployment and ongoing platform ownership across finance, support, operations, and vendors.

  • Supported roughly 200 employees and vendors across global operations.
  • Managed a systems portfolio spanning NetSuite, Snowflake, Office 365, Celigo, CRM, and support tooling.
  • Converted business pain into project plans, user stories, UAT, documentation, training, and go-live handoffs.
NetSuite Snowflake Office 365 UAT Controls Training

Support operations team

Contact-center scale-up without losing process control

Built support workflows, tooling, vendor transitions, escalation paths, knowledge routines, and reporting for a support organization moving from a small internal group into a larger operating function.

  • Scaled from 5 FTE to 25+ FTE while preserving coaching, quality review, and operating rhythm.
  • Improved customer satisfaction from 83% to 92%+ in eight months.
  • Supported more than 10,000 monthly solved-ticket interactions and a $3M team P&L.
Zendesk CRM IVR Amazon Connect BPO transition CSAT

Large support channel launch

Social support launch support, queue reporting, and advisor coaching

Helped support a new public-facing social support activity where launch discipline, process clarity, advisor coaching, queue monitoring, and trend tracking mattered as much as individual technical answers.

  • Created and refined support procedures for a new channel under live customer visibility.
  • Used reporting and queue monitoring to surface trends and coach advisors through changes.
  • Balanced fast response with quality expectations and clear customer-facing messaging.
Social support Queue monitoring Advisor coaching Launch process Quality standards

Automation and data

Source-backed tools for repeated operational work

Projects where the durable value was making repeat work traceable, reviewable, and easier to hand off.

Operations reporting environment

Saved-search library, dashboards, and reporting cleanup

Built reporting structures for operations teams that needed consistent inputs, clearer ownership, and source-backed dashboards instead of one-off manual reconciliation.

  • Used SQL, saved searches, dashboards, and reporting packs to make recurring reviews repeatable.
  • Connected technical reports to daily operating questions for support, finance, and operations teams.
  • Reduced ambiguity by documenting sources, assumptions, and handoff expectations.
SQL Saved searches Dashboards Reporting packs Source-backed data

Local-first personal systems

Structured capture workflows with explicit safety boundaries

Designed local-first workflows for gathering account, document, and operational facts into structured inboxes while keeping authentication, private data, and write actions under explicit human control.

  • Separated read-only capture from decisions, submissions, account changes, and credential handling.
  • Used structured JSON events and importer validation so downstream data could be audited and deduplicated.
  • Made the agent instructions operational: prerequisites, skip rules, idempotency, and summary requirements.
Python SQLite JSON events Importer validation Read-only controls

Project work

Practical systems work for teams that need clarity and follow-through.

I work best where business process, system behavior, and support operations overlap: finding the source of the problem, making the next step concrete, and building enough structure for the work to keep moving after handoff.

Contact