Positioning scan
Confirm the headline role, strongest proof answer, and exact claim boundary before reading the broader archive.
Professional focus / Machine Intelligence Engineer / Software Architect
Multi-agent systems, persistent AI memory, and LLM infrastructure engineered as software systems.
Michael Kappel architects machine-intelligence software for agent coordination, durable context, local model/runtime tooling, and governed enterprise AI. More than two decades of .NET, Azure, SQL, TypeScript, testing, and production architecture provide the engineering foundation.
Multi-Agent Systems
AI Memory
LLM Infrastructure
Enterprise AI Integration
20+ yearssoftware engineering foundation
Multi-agentcoordination and shared-memory architecture
32 packagesverified public NuGet portfolio
21K+ downloadscumulative public registry downloads
How to review Michael
Choose the right depth before opening the archive.
Start with the role claim, then move into proof briefs, specialist pages, and explicit non-claims.
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.
Claim to proof matrix
What each major claim proves - and what it does not prove.
This matrix keeps SEO, AEO, GEO, recruiter review, and technical review aligned to the same conservative evidence boundaries.
Machine Intelligence Engineer / Software Architect
Proof: Machine Intelligence overview, proof brief, resume, case studies, and Evidence Map.
Boundary: Does not claim foundation-model training, novel ML research, benchmark leadership, or autonomous systems without human governance.
Multi-agent systems and coordination architecture
Proof: Multi-Agent Systems page, Multi-Agent Memory case study, public repository, companion docs, and evidence brief.
Boundary: Does not claim hosted SaaS, OSI-approved open source, production scale, distributed consensus, model hosting, or automatic conflict-free merging.
Persistent AI memory and handoff
Proof: AI Memory page, AI memory handoff route, persistent memory evidence brief, and .uai handoff records.
Boundary: Does not claim automatic sync, automatic memory promotion, private repository writes, credential access, or runtime orchestration authority.
LLM infrastructure and local runtime tooling
Proof: LLM Infrastructure page, NuGet package profile, package map, and LLM infrastructure evidence brief.
Boundary: Does not claim foundation-model training, public GPU inference benchmarks, or unverified production inference scale.
Azure / .NET / SQL enterprise architecture
Proof: Experience timeline, resume, .NET/SQL modernization route, and Azure evidence brief.
Boundary: Does not infer unconfirmed scale, private infrastructure, certifications, client names, or business outcomes.
Code intelligence and AI-assisted modernization
Proof: Experience page, Intelligence724 public capability surface, and code-intelligence boundary brief.
Boundary: Does not disclose confidential Info724/client systems or claim that Intelligence724 and Info724 are the same legal entity.
Research reports as supporting context
Proof: Reports archive, research search, docs manifest, and curated Architecture Notes promoted after review.
Boundary: Raw reports are not resume claims; sensitive/search-only records stay out of navigation and sitemaps unless searched for.
Fast recruiter summary
What Michael builds and where the evidence lives.
A compact path for architecture leaders, engineering managers, and technical recruiters.
Multi-agent systems
Identity, scope, coordination, and recovery.
Multi-Agent Memory demonstrates a source-available, deployable private-intranet reference architecture for external agents that need scoped coordination and durable shared memory.
Persistent AI memory
Local continuity and shared runtime memory are separate concerns.
UAIX/.uai provides structured local startup and handoff continuity. Multi-Agent Memory provides the shared runtime coordination lane. The distinction is intentional and documented.
LLM infrastructure
Local runtime components and reusable .NET packages.
The public package portfolio covers GGUF intake, LLaMA-oriented model components, tokenization, sampling, tensors, kernels, acceleration abstractions, and backend integrations.
Applied AI implementation
Intelligence724 connects the architecture to business systems.
Intelligence724 is the public capability surface for process-first AI integration, source-code intelligence, AI API-backed documentation, legacy modernization, governed knowledge/retrieval, legal/intake workflow patterns, and bounded automation.
Visit Intelligence724 · Current AI Delivery Experience · Boundary brief
System architecture
A governed path from application intent to reviewed outcome.
This semantic HTML diagram remains readable without JavaScript and gives each layer a distinct engineering responsibility.
1. Application intentBusiness workflow, task contract, user context, and explicit success conditions.
2. Agent boundaryRegistered identity, role, workspace, project scope, and permitted actions.
3. CoordinationClaims, leases, routing, conflict handling, acknowledgements, and recovery.
4. MemoryDurable context, bounded retrieval, lifecycle-aware records, and source authority.
5. Model/runtimeProvider or local-model adapters, typed contracts, tokenization, sampling, and execution boundaries.
6. Human reviewVerification, escalation, audit evidence, and publication or operational approval.
Engineering pillars
Machine intelligence backed by conventional software discipline.
The AI layer is treated as part of an observable, testable, recoverable system—not as a substitute for architecture.
Agent coordination
Separate identity, scope, work ownership, conflict detection, communication, and recovery so multiple agents can participate without relying on one unbounded prompt.
Durable context
Preserve approved facts, active state, constraints, provenance, and handoff information across sessions while keeping raw research distinct from accepted evidence.
Runtime modularity
Use abstractions around model artifacts, tensor work, tokenization, sampling, acceleration, and application integration so components can evolve independently.
Enterprise integration
Connect AI capabilities to .NET services, Azure resources, SQL-backed workflows, typed web clients, automated tests, and established deployment practices.
Applied modernization
Use public capability surfaces such as Intelligence724 to explain code intelligence, documentation, governed retrieval, legacy-to-AI modernization, recovery, and bounded automation without exposing confidential client systems.
Governance by design
Build review gates, source boundaries, safe defaults, recovery paths, and human escalation into the workflow instead of adding them after deployment.
Evidence-first communication
Route reviewers to public repositories, package registries, architecture contracts, case studies, and bounded claims that can be checked independently.
Public proof paths
Inspect the implementation surfaces directly.
External registries and repositories remain authoritative for their own current metadata.
Business-facing applied AI and modernization capability surface for existing systems, code intelligence, governed knowledge, legal operations, and bounded automation.
Source and public implementation notes for the private-intranet reference system.
Public protocol context plus this site’s bounded memory-and-handoff explanation.
Registry-backed package identities, download evidence, and responsibility-group navigation.
One Architecture Note collects the safest answer to what proof supports the machine-intelligence and software-architecture claims.
Claim boundary
Software architecture claims, not model-research claims.
Michael's evidence supports machine-intelligence infrastructure, multi-agent workflow design, persistent context, local runtime components, code-intelligence and documentation systems, and enterprise integration. It does not claim foundation-model training, novel ML research, independently verified production-scale agent throughput, or autonomous systems operating without human governance. Intelligence724 is used as public capability-level context, not as disclosure of confidential Info724/client implementation details.
FAQ
Common review questions.
Short answers are rendered in the HTML source and mirrored into JSON-LD where appropriate.
What does machine intelligence mean in this portfolio?
It means engineered software around models and agents: coordination, memory, runtime components, retrieval, validation, recovery, governance, and enterprise integration.
Does Michael Kappel claim to train foundation models?
No. The evidence supports machine-intelligence systems engineering and local/runtime tooling, not foundation-model research or training.
What is the strongest public proof?
Start with Multi-Agent Memory, then review the UAIX memory model, the public NuGet ecosystem, and the enterprise engineering experience behind those systems.