Need multi-agent proof?
Open Multi-Agent Memory first for identity, scopes, durable reviewed memory, rooms, messages, claims, leases, and human verification boundaries.
Open Multi-Agent MemoryMachine intelligence and software architecture case studies
Multi-agent coordination, AI memory, LLM infrastructure, and enterprise engineering.
Seven evidence-bounded case studies translate public implementations, platforms, and resume-backed work into constraints, architecture decisions, validation methods, and explicit limits a technical reviewer can assess.
7case studies
3review lanes
5approved or confirmed
100%explicit claim boundaries
Case selector for reviewers
Which case should I open first?
Use the case library as evidence triage: pick the first case by the hiring question, then continue only if deeper architecture detail is needed.
Need AI memory governance?
Open UAIX and LLMWikis cases for handoff contracts, trust labels, retrieval boundaries, provenance, and human approval gates.
Open UAIX caseNeed enterprise modernization?
Open the enterprise modernization and Angular/Python cases to connect machine-intelligence work back to .NET, SQL, TypeScript, testing, and delivery discipline.
Open enterprise laneNeed evidence boundaries?
Every case separates public proof, design-review evidence, and explicit non-claims so reviewers do not infer confidential details or unsupported production scale.
Claim-boundary matrixHow to review Michael
Choose the right depth before opening the archive.
Use this path to decide whether a quick hiring-manager read is enough or a deeper architecture review is warranted.
Positioning scan
Confirm the headline role, strongest proof answer, and exact claim boundary before reading the broader archive.
Recruiter path
Use the concise resume, current contact path, and role-level experience timeline to qualify fit quickly.
Hiring-manager path
Move from role fit into selected projects and case studies, then inspect the Evidence Map if a claim needs traceability.
Technical architecture path
Inspect Multi-Agent Memory, persistent AI memory, local LLM infrastructure, package evidence, validation, and non-claims.
Machine-readable review path
Use this path for AI agents, answer engines, ATS tools, and structured review without treating hidden files as public claims.
Machine-intelligence claim path
Answer the core proof question directly, then branch into multi-agent systems, AI memory, LLM infrastructure, and Intelligence724 boundaries.
Azure / .NET / SQL path
Confirm the enterprise engineering foundation with role-level Azure usage, .NET/SQL modernization, and conservative evidence boundaries.
How to evaluate the work
A four-step technical review.
Each case separates public evidence from private implementation details, so architecture judgment can be reviewed without invented metrics or confidential disclosure.
- 01Select a lane
Start with the technical problem closest to the open role.
- 02Read the constraint
See what could regress, drift, or become unsafe.
- 03Check validation
Review how the implementation claim is tested or bounded.
- 04Open the evidence
Follow the detail route into resume, docs, and public proof.
Featured public source-available proof
Public Source Available Reference Implementation
Enterprise Validated
Multi-Agent Memory Coordination System
A deployable private-intranet, source-available reference implementation for coordinating external AI agents through registered identities and scoped grants, persistent reviewed memory, meeting and current-message routing, acknowledgements, and hash/CAS/lease controls for simultaneous local .uai edits.
External AI agents working on the same initiative can lose durable context, duplicate work, overwrite local continuity files, or act outside their intended scope when identity, routing, memory, and edit coordination remain implicit.
Make agent identity and scope explicit before accepting shared coordination or memory operations.
Review the public source and companion documentation, run the repository validation suite, and trace identity, scope, memory, message, acknowledgement, hash/CAS, claim, lease, and persistence contracts while checking every explicit non-claim.
.NET 9
C#
HTTP APIs
JSON
SQLite
MySQL
MariaDB
Hash and CAS Coordination
Bounded Leases
Case study library
Compare the supporting evidence lanes.
Cards provide the challenge, first architecture move, validation method, proof signal, and public boundary before asking the reviewer to open a detail page.
AI systems and knowledge governance
Continuity, trust state, and safe agent boundaries.
Public-platform and design-review cases for AI memory, durable knowledge, discovery controls, and human approval boundaries.
Public Platform
Confirmed Experience
UAIX AI Memory Package Wizard
A public-platform case study for AI memory, file handoff, project handoff, and compact AI handoff state mapped to long-term Architecture Notes research.
- Challenge
- AI-assisted engineering loses continuity when source files, decisions, constraints, and current state are scattered across chats, one-off notes, and non-reviewable prompts.
- Architecture move
- Model handoff files as explicit read-order and approval-boundary artifacts.
- Validation
- Validate JSON exports, mirror alignment, package read order, and generated-needs-review status for safety anchors before release packaging.
What this demonstrates
- AI memory as a structured engineering artifact
- Agent handoff with reviewable state
UAIX
AI handoff
AI Memory
Project Handoff
File Handoff
Markdown
Public Platform
Confirmed Experience
LLMWikis Trust-Labeled Knowledge
A public-platform case study for durable AI-ready knowledge systems, trust labels, source boundaries, governance, and retrieval-friendly content design.
- Challenge
- LLM-facing documentation can become a flat pile of pages with no source state, claim boundary, or reviewer route for humans and AI evaluators.
- Architecture move
- Frame knowledge entries around trust labels, route provenance, and claim boundaries.
- Validation
- Check JSON parse, route index consistency, noindex boundary, and evidence labels before promoting research content into public claims.
What this demonstrates
- Knowledge systems designed for humans and agents
- Governance-aware content packaging
LLM Wiki
Markdown
frontmatter
metadata
trust labels
RAG
Prototype / Technical Review
Design Review
NeuralWikis / LocalEndpoint AI Boundaries
A review-stage prototype case study for Python, MySQL, cognitive packet review, passive validation, and no-execution safety boundaries.
- Challenge
- AI agents need useful discovery and memory exchange without broad tool authority, private-network probing, or unreviewed packet adoption.
- Architecture move
- Describe hostile-context and zero-blind-import concepts as reviewable boundaries rather than certification claims.
- Validation
- Validate no private helper mirror, public discovery boundary language, JSON manifests, and noindex utility inventory.
What this demonstrates
- Security-boundary thinking for AI agents
- Python-oriented metadata validation patterns
Python
MySQL
TypeScript
JSON
metadata validation
passive validation
Enterprise software and evidence UX
The engineering foundation remains visible and reviewable.
Resume-backed modernization, Angular/RxJS service boundaries, and browser-local evidence routing show the .NET, SQL, TypeScript, testing, and delivery foundation beneath the machine-intelligence work.
Resume Backed Professional Pattern
Enterprise Validated
Enterprise Modernization and Parity Validation
A resume-backed case study for modernizing legacy Web Forms, Classic ASP, and SQL-heavy systems while preserving business behavior through comparison screens, generated scenarios, and test coverage.
- Challenge
- Legacy business systems need modernization while preserving behavior embedded in SQL, forms, and old workflow assumptions.
- Architecture move
- Capture legacy behavior and SQL-side business rules before changing implementation boundaries.
- Validation
- Use behavioral comparison screens, generated scenario tests, automated unit/integration testing, and source review before treating new code as behavior-compatible.
What this demonstrates
- Legacy modernization planning around behavior preservation
- SQL-heavy business-rule translation into reviewable architecture
ASP.NET Core
C#
SQL Server
T-SQL
stored procedures
TypeScript
Reference Architecture
Design Review
Angular / RxJS / Python Service Boundary Architecture
Enterprise Angular/RxJS architecture example with TypeScript DTOs, RxJS data access, Python service boundaries, SSR/SSG, hydration, testing, and security.
- Challenge
- Public portfolios and commercial web applications need rich interactivity without sacrificing crawlability, accessibility, or clear API/review boundaries.
- Architecture move
- Present TypeScript and Angular experience through source-backed route and resume evidence.
- Validation
- Keep professional claims tied to resume lines; label reference architecture separately; validate JavaScript syntax and responsive component behavior.
What this demonstrates
- Enterprise Angular/RxJS architecture reasoning
- SEO-first rendering patterns
Modern Angular
RxJS
TypeScript
SSR
SSG
Hydration
Public Platform
Enterprise Validated
MikeKappel.com Capability Resolver
A public-platform case study for browser-local job matching, alias normalization, evidence cards, proof pages, and claim-boundary warnings.
- Challenge
- Recruiters and technical reviewers need a fast way to map role language to evidence without fake fit percentages or hidden ATS-style scoring.
- Architecture move
- Keep the matcher optional rather than the dominant homepage surface.
- Validation
- Run JavaScript syntax checks, component QA for menu/tool behavior, and text-extraction scans for concatenated card text.
What this demonstrates
- Portfolio as proof system
- Evidence-first recruiter workflow
JavaScript
WordPress
REST
JSON
HTML
CSS
Evidence standard
Useful detail without unsafe disclosure.
The collection is designed for recruiters, hiring managers, technical interviewers, search systems, and AI review tools reading the same visible evidence.
Know what is confirmed.
Approved, user-confirmed, and design-review labels remain visible before stronger claims are made.
See how claims are checked.
Tests, parity screens, JSON validation, route checks, and public evidence paths are named instead of implied.
Know what is not claimed.
Private code, client data, unsupported metrics, autonomous authority, and unverified production behavior remain explicitly out of scope.
Case study FAQ
Questions technical reviewers ask.
These visible answers match the route's FAQ structured data.
What does each software architecture case study include?
Each case study names the business problem, technical constraint, architecture strategy, validation method, evidence status, technologies, current result, and an explicit boundary for what is not publicly claimed.
Which case study should a recruiter or hiring manager review first?
Start with Enterprise Modernization and Parity Validation for .NET and SQL Server roles, the AI systems lane for memory and knowledge-governance work, or the frontend lane for Angular, RxJS, TypeScript, and evidence UX.
How do these case studies keep technical claims trustworthy?
The collection exposes source status, validation methods, public evidence links, and claim boundaries. Private code, client data, unsupported metrics, and unverified production behavior remain out of scope.