1‐ Before vs After AI - magicplatforms/ai-workflows GitHub Wiki
Below is an updated Azure DevOps Wiki page that eliminates the “got ‘note’” parse error and other potential rendering issues. Key fixes:
- Every label that appears in a message is now declared first with
participant(Mermaid is strict about this). - Reserved keywords such as
Notehave been renamed (e.g.,ResearchNote,AssessmentReport). - No spaces are used in raw participant IDs; readable labels are supplied with
as.
Copy everything between the horizontal rules into a new Wiki page.
AI‑Driven Workflow Transformation
[[TOC]]
[!NOTE] Mermaid diagrams render automatically once the page is saved or previewed (be sure the preview feature “Render Mermaid diagrams in Wiki and Markdown” is enabled).
Sales – Before vs After AI
sequenceDiagram
title Sales – Before vs After AI
%% participants
participant SDR
participant AE
participant LinkedIn
participant Prospect
participant CRM
participant IntentDB
participant AI_Agent as AI Agent
participant Calendar
participant Win_Coach as Win Coach
%% flows
alt **Before AI**
SDR->>LinkedIn: Scrape leads
SDR->>Prospect: Draft & send emails
SDR->>CRM: Log calls
AE->>CRM: Update opps nightly
else **After AI**
AI_Agent->>IntentDB: Mine intent signals
AI_Agent->>Prospect: Auto‑personalise outreach
AI_Agent->>Calendar: Schedule call
AE->>Win_Coach: Realtime win‑probability tips
CRM-->>AE: Forecast auto‑roll‑up
end
Human Resources – Before vs After AI
sequenceDiagram
title HR – Before vs After AI
participant Recruiter
participant Candidate
participant HRBP
participant ResumeDB
participant Calendar
participant GenAI_Ranker as GenAI Ranker
participant Chatbot
participant HR_Analytics as HR Analytics
participant Panel
alt **Before AI**
Recruiter->>ResumeDB: Screen 200 résumés
Recruiter->>Candidate: Phone screens
Recruiter->>Calendar: Block interview slots
HRBP->>HR_Analytics: Build turnover deck monthly
else **After AI**
GenAI_Ranker->>ResumeDB: Rank résumés
Chatbot->>Candidate: Run first interview
Calendar->>Panel: Auto‑book panels
HR_Analytics-->>HRBP: Live attrition heat‑map
HRBP->>HR_Analytics: Push‑button stay‑bonus scenario
end
Legal – Before vs After AI
sequenceDiagram
title Legal – Before vs After AI
participant Associate
participant Paralegal
participant Westlaw
participant Client
participant Research_LLM as Research LLM
participant ContractAI
participant Docs
participant Team
alt **Before AI**
Associate->>Westlaw: Hunt precedent
Associate->>Client: Draft memo & redline contract
Paralegal->>Docs: Bates‑stamp discovery
else **After AI**
Research_LLM->>Westlaw: Search & summarise precedent
ContractAI->>Client: Mark risk clauses & draft NDA
Paralegal->>Docs: Validate AI‑coded discovery
Paralegal->>Team: Prep strategy brief
end
Finance / Banking – Before vs After AI
sequenceDiagram
title Finance / Banking – Before vs After AI
participant Analyst
participant Excel
participant Compliance
participant MarketAI
participant Surveillance_AI as Surveillance AI
participant DataFeeds
participant ResearchNote
participant Trades
participant FilingsDB
alt **Before AI**
Analyst->>DataFeeds: Copy data to Excel
Analyst->>Excel: Build valuation model
Analyst->>ResearchNote: Write 8‑page report
Compliance->>Trades: Hand‑check overnight
else **After AI**
MarketAI->>FilingsDB: Ingest filings & generate model
MarketAI-->>ResearchNote: Draft equity research note
Analyst->>MarketAI: Stress‑test assumptions
Surveillance_AI-->>Trades: Pre‑clear in seconds
end
Education – Before vs After AI
sequenceDiagram
title Education – Before vs After AI
participant Teacher
participant Student
participant Parent
participant Grader
participant Lesson_Generator as Lesson Generator
participant TutorAI
participant Analytics
participant LessonPlan
participant Classroom
alt **Before AI**
Teacher->>LessonPlan: Craft lesson plan
Teacher->>Grader: Manually grade papers
Teacher->>Parent: Send emails
else **After AI**
Lesson_Generator-->>Teacher: Differentiated lesson plan
TutorAI->>Student: Instant feedback
Analytics-->>Teacher: Performance dashboards
Teacher->>Classroom: Socratic discussion
Teacher-->>Parent: 30‑sec video update
end
Technology / Software Dev – Before vs After AI
sequenceDiagram
title Software Dev – Before vs After AI
participant Engineer
participant Repo
participant CI
participant Copilot
participant BuildHelp_LLM as BuildHelp LLM
participant CodeBase
participant Prod
alt **Before AI**
Engineer->>Repo: Design API & write boilerplate
Repo->>CI: Wait for build & tests
CI-->>Engineer: Code review feedback
else **After AI**
Copilot-->>Repo: 60–80 % scaffolding & autotests
Engineer->>CodeBase: Focus on edge cases & architecture
CI->>BuildHelp_LLM: Build errors
BuildHelp_LLM-->>Engineer: Explain fixes in chat
CI-->>Prod: Release cadence halves
end
Healthcare – Before vs After AI
sequenceDiagram
title Healthcare – Before vs After AI
participant Physician
participant Patient
participant Guidelines
participant Coder
participant Ambient_Scribe as Ambient Scribe
participant ClinicalLLM
participant EHR
participant ClaimDoc
alt **Before AI**
Physician->>EHR: Type notes
Physician->>Patient: Dictate letters
Physician->>Guidelines: Search care guidelines
Coder->>ClaimDoc: Assign CPT codes
else **After AI**
Ambient_Scribe-->>EHR: Auto‑generate note
ClinicalLLM->>Physician: Diagnostic suggestions & orders
Coder->>ClaimDoc: Audit AI‑assigned codes
Physician->>Patient: Spend extra time counselling
end
Supply‑Chain / Ops – Before vs After AI
sequenceDiagram
title Supply‑Chain / Ops – Before vs After AI
participant Planner
participant ERP
participant Supplier
participant Demand_Model as Demand Model
participant GenAI
participant Digital_Twin as Digital Twin
participant Routing_Engine as Routing Engine
participant Dispatcher
participant Routes
participant Carrier
participant Exceptions
alt **Before AI**
Planner->>ERP: Export data & tweak forecast
Planner->>Supplier: Call for updates
Dispatcher->>Routes: Build route sheet
else **After AI**
Demand_Model-->>Planner: Demand forecast
GenAI->>Supplier: Draft PO & negotiation email
Digital_Twin-->>Planner: Network simulation
Routing_Engine-->>Carrier: Issue loads autonomously
Planner->>Exceptions: Handle only anomalies
end
Insurance (P&C / Life) – Before vs After AI
sequenceDiagram
title Insurance – Before vs After AI
participant Adjuster
participant Underwriter
participant Claim
participant CV_Model as CV Model
participant GenAI
participant FraudAI
participant PartsDB
participant AssessmentReport
participant ClaimPhotos
participant RiskAI
participant RiskTables
alt **Before AI**
Adjuster->>Claim: Read claim & request photos
Adjuster->>PartsDB: Price parts
Adjuster->>AssessmentReport: Write assessment
Underwriter->>RiskTables: Crunch risk tables
else **After AI**
CV_Model->>ClaimPhotos: Score damage from images
GenAI-->>Claim: Draft settlement
Underwriter->>RiskAI: Review AI risk score
Underwriter->>RiskAI: Tune appetite
FraudAI-->>Claim: Flag anomalies in real‑time
end
End of page