01 — Our Work

Production systems, shipped.

Our client engagements are covered by non-disclosure agreements, so the case studies below are anonymized. On a call we walk through the architecture, the tradeoffs and the delivery timelines in detail — including systems comparable to whatever you're trying to get into production.

Book a Call with a Builder
— What we build

Four practices, one engineering organization.

01 / Software Engineering

Software Engineering

Product engineering and platform builds for commerce, payments, supply chain, and customer experience systems. Event-driven architectures, API platforms, and developer experience.

MicroservicesKafka · EventBridgeComposable CommercePlatform Engineering
02 / Applied AI

Applied AI

From foundation-model selection to production deployment: RAG architectures, agentic workflows, and AI-augmented operations grounded in enterprise data — with the governance to ship them safely.

LLM PlatformsRAG · AgentsMLOpsAI Governance
03 / Cloud Modernization

Cloud Modernization

Landing zone design, workload migration, FinOps, and multi-cloud governance. We modernize what's worth modernizing — and decommission what isn't — with a clear-eyed business case.

AWS · Azure · GCPFinOpsMigrationLanding Zones
04 / Cloud Security

Cloud Security

Zero-trust architecture, identity modernization, CSPM, and incident response. Built for organizations where breach blast-radius is measured in regulatory exposure, not just dollars.

Zero TrustCSPMIAMDevSecOps
— Case studies

What recent engagements produced.

Anonymized at our clients' request. Industry, problem, what we built, and the business outcome.

Applied AI

Higher Education — Top-tier research university (40,000+ students)

Enterprise agent adoption roadmap and student career navigation

The problem

Fragmented internship tracking, manual advisor bottlenecks and high drop-off during certification and career program enrollment. Leadership needed an enterprise-wide agent adoption roadmap, proven first in a low-risk lighthouse deployment that validated technology, governance, FERPA compliance and adoption.

What we built

  • Partnered with university leadership on an enterprise agent use-case adoption roadmap and governance framework.
  • Selected the School of Continuing Education as the lighthouse school to prove the deployment model.
  • Engineered a production Gemini Enterprise agent that manages internship tracking, monitors milestones and guides students through internship and professional certification enrollment.
  • Integrated the student information system and career portals with FERPA-compliant privacy boundaries and automated escalation to human advisors.

Timeline — 6-week lighthouse pilot, followed by a 14-week enterprise expansion.

The outcome

73%
Reduction in administrative enrollment review time
4.2x
Increase in on-time student milestone completions in the first term
68%
Self-serve engagement among continuing education learners
8
Additional colleges approved for phased rollout
Cloud Modernization

Financial Services — Regional commercial and retail bank ($15B+ AUM)

Cloud modernization and a personalized banking data platform

The problem

Rising colocation costs and legacy on-premise infrastructure kept customer data trapped in siloed systems. Client services and advisory teams worked from 36-hour batch reports, limiting their ability to deliver timely, personalized financial services.

What we built

  • Consolidated and decommissioned the on-premise data center, migrating client services applications and supporting infrastructure to Microsoft Azure.
  • Delivered a modern enterprise data warehouse with near-real-time pipelines unifying core banking, CRM and account servicing data.
  • Built visualization and reporting for client services and operations teams: 360-degree customer views, portfolio insights and next-best-action alerts.

Timeline — 18-week phased migration and data warehouse build, with zero downtime for customer-facing applications.

The outcome

38%
Reduction in annual infrastructure run rate
92%
Reduction in reporting latency — 36 hours to under 15 minutes
24%
Increase in personalized product and advisory conversion
0
Material findings in the pre-cutover compliance audit
Applied AI

Advertising & Marketing — Multi-brand campaign performance agency

Agentic workflow orchestration for campaign operations

The problem

Campaign teams lost time switching between CRM, email and external data tools. Account managers spent over 60% of their hours synthesizing data, drafting segment variants and reconciling audience records, capping capacity and slowing launches.

What we built

  • Defined the AI operationalization, workflow automation and marketing operations strategy, with a controlled path from pilot to organization-wide rollout.
  • Engineered an agent orchestration layer with OpenAI ChatGPT as the orchestrator across the existing campaign workflow.
  • Built Model Context Protocol (MCP) servers connecting the orchestrator to the CRM, email service, third-party data aggregators and enrichment APIs, with centralized credential and rate-limit governance.
  • Automated lead enrichment, campaign brief synthesis and copy variation with human-in-the-loop brand review.

Timeline — 4-week pilot on high-volume campaigns, followed by a 10-week production rollout across all account teams.

The outcome

65%
Faster campaign launch — 5 business days to under 36 hours
40+
Hours saved per campaign manager each month
31%
Increase in engagement and conversion rates
2.8x
Campaign throughput with no added headcount
Applied AI

Enterprise Software — Venture-backed HRTech SaaS platform

Real-time multi-modal AI interview platform

The problem

A stretched founding engineering team, and enterprise hiring customers overwhelmed by manual first-round screening — multi-week cycles, recruiter fatigue and candidate drop-off. They needed a reliable audio-and-text interview agent that met enterprise latency, scoring-consistency and bias-mitigation standards.

What we built

  • Embedded senior AI and cloud engineers to build a real-time, multi-modal interview agent on LangGraph.
  • Orchestrated audio input, structured reasoning and evaluation, and voice output in a single graph.
  • Implemented rubric-based evaluation with guardrails and human-in-the-loop review queues.
  • Deployed on Azure Kubernetes Service with CI/CD through GitHub and Azure DevOps, plus latency monitoring and tracing.

Timeline — 10 weeks from architecture validation to live production.

The outcome

75%
Reduction in first-round screening time — 12 days to under 48 hours
3
Modalities orchestrated with sub-second response latency
<2%
Exception rate across concurrent live interview sessions
3.5x
Hiring pipeline capacity without added recruiter headcount
— Related services

How the work is delivered.

Ready to get back to building?

Tell us about the engagement. We typically respond within one business day with a named Builder who can talk substance — not a generic sales pitch.

Book a Call with a Builder

Prefer email? info@inthe.cloud

What happens next

  • 30-minute builder-led call
  • No generic sales pitch
  • Architecture, feasibility and constraints
  • A recommended next step