Case study structure used across customer engagements.
Case studies and artifacts from forward-deployed AI engineering engagements.
git clone https://github.com/cmangun/field-deployed-engineer-showcase.git
cd field-deployed-engineer-showcase
cat case-studies/pfizer-cocreate.mdOr view online: github.com/cmangun/field-deployed-engineer-showcase
These case studies demonstrate:
- 35% faster compliance reviews (ML automation)
- 98% reduction in knowledge discovery time (RAG)
- Zero FDA audit findings (pipeline migration)
| Client | Project | Outcome |
|---|---|---|
| Pfizer | RAG Knowledge Platform | 98% faster discovery, 35% faster compliance |
| Abbott | ML Pipeline Migration | Zero FDA findings, 6mo → 3wk deployment |
| IPG Health | Platform Consolidation | $51M portfolio, 99.9% uptime |
Each case study follows the FDE narrative structure:
┌─────────────────────────────────────────────────────────────┐
│ 1. CONTEXT │
│ Industry, constraints, stakeholders │
├─────────────────────────────────────────────────────────────┤
│ 2. PROBLEM │
│ What was failing, slow, risky, or non-compliant │
├─────────────────────────────────────────────────────────────┤
│ 3. APPROACH │
│ Discovery → Design → Build → Deploy → Stabilize │
├─────────────────────────────────────────────────────────────┤
│ 4. DEPLOYMENT & STABILIZATION │
│ Runbooks, monitoring, incident handling │
├─────────────────────────────────────────────────────────────┤
│ 5. OUTCOMES │
│ Latency, throughput, adoption, business impact │
├─────────────────────────────────────────────────────────────┤
│ 6. WHAT I WOULD PRODUCTIZE │
│ Patterns extracted for reuse │
└─────────────────────────────────────────────────────────────┘
| Artifact | Purpose |
|---|---|
runbook-example.md |
Operational runbook template |
postmortem-example.md |
Incident postmortem format |
adr-example.md |
Architecture decision record |
| Repo | Demonstrates |
|---|---|
| deployable-ai-agents | Agent patterns from engagements |
| enterprise-llm-integration | LLM governance from Pfizer work |
| regulated-data-pipelines | ETL patterns from Abbott work |
- Add Medtronic case study
- Add Amgen case study
- Add architecture diagrams (Mermaid)
- Add engagement timeline visuals
- Add lessons learned section
MIT © Christopher Mangun
Portfolio: field-deployed-engineer.vercel.app
