About All levels 4 minute read Updated 2026-06-26 UTC

Contact and About Michael Kappel

Direct contact information and a technical biography for Michael Kappel, the public creator contact behind ModelBreeder.com.

Research statusPublic crawl synthesis Publication statePublished Reviewed byMichael Kappel Source reports1

Direct contact

Michael Kappel is the public creator contact for ModelBreeder.com. His public portfolio presents him as a Senior Software Engineer and Software Architect focused on AI-assisted engineering, .NET, SQL Server, TypeScript, enterprise modernization, and source-governed project memory.

RouteDetail
Emailmike@ns12.com
Phone(630) 362-7576
Compact contact hubmjk.tel
Professional siteMikeKappel.com
Resumemikekappel.com/resume
GitHubgithub.com/MichaelKappel
NuGetnuget.org/profiles/Michael.Kappel

The concise public contact hub also lists Signal as Mike.7576 and provides alternate phone routes for direct reachability.

Why this background fits ModelBreeder

ModelBreeder.com is about adaptive systems that can change without losing accountability. That requires lineage, contracts, tests, release gates, local runtime awareness, rollback discipline, and durable memory. Michael's public work history aligns with those requirements because much of enterprise modernization is the practical art of preserving behavior while replacing brittle internal structure.

ModelBreeder concernMatching engineering signal
Lineage and regression controlLegacy modernization, parity validation, comparison screens, generated scenarios, unit tests, integration tests, and review before behavior-compatible replacement.
Typed capability contractsASP.NET/Core APIs, TypeScript DTOs, Angular/RxJS patterns, Web API boundaries, Microsoft Graph API work, and explicit browser/API contracts.
Local-first AI infrastructurePublic package work around UAI/UAIX, local memory stores, adaptive interoperability, local LLM runtime abstractions, GGUF/runtime surfaces, tokenization, tensors, sampling, CPU kernels, and acceleration boundaries.
AI memory and handoffAgent File Handoff, project handoff, source-governed prompt architecture, semantic search, durable review notes, and machine-readable continuation context.
Operational adoptionMentoring, Agile practices, Azure DevOps workflows, CI/CD context, code review, SOLID/design-pattern training, and maintainability-first architecture decisions.

Technical biography

Michael's career spans more than two decades of enterprise software delivery. His public portfolio emphasizes business-critical modernization: stabilizing Web Forms, Classic ASP, SQL-heavy systems, undocumented business rules, service seams, data migration paths, reporting surfaces, and reviewable evidence before treating a rebuilt system as equivalent to the old one.

At Info724 / Insurance 724, his public experience summary includes AI-assisted engineering workflows using OpenAI-compatible APIs, LM Studio, prompt systems, semantic search, and reviewed AI output for source analysis, documentation, and legacy-to-modern code translation. It also lists Angular/TypeScript AI documentation review work, ASP.NET Core/C#/TypeScript/EF Core/Razor Pages contract-management work, SQL Server/MySQL/AWS-backed workflows, Power BI/DAX reporting, and mentoring around testing and architecture.

His prior roles add breadth across collaboration workflows, document and file handling, Azure Blob Storage, Microsoft Graph API integration, Azure DevOps, QUnit, Angular/RxJS, corporate tax workflows, Classic ASP modernization, ASP.NET Core Web API, logistics and transportation-management architecture, secure layered enterprise systems, WPF/Prism banking software, ERP/e-commerce modernization, public-sector airport systems, and early commercial web production.

Public package and code evidence

Michael's public NuGet profile is a strong .NET signal for this project. The profile lists packages around UAI/UAIX message contracts, AI memory, project handoff, adaptive interoperability, ErrorNotifier logging, Talisman agent surfaces, local LLM runtime components, GGUF/runtime packages, tokenization, tensors, sampling, CPU kernels, and acceleration contracts.

Those packages matter to ModelBreeder because adaptive AI systems need more than theory. They need typed boundaries, local memory, package identity, runtime diagnostics, portable artifacts, review checkpoints, and repeatable evidence.

His GitHub profile adds public code context, including C# repositories for Fibonacci, repository patterns, cryptography, SpeedCube/Rubik's Cube work, and Mailjet API wrapper work. The public profile also identifies Long Term Software Solutions, Chicago IL, MichaelKappel.com, and the @MichaelKappel social handle.

Contact use cases

Use this route for technical discussion around model breeding, UAI/UAIX memory packages, local model runtime boundaries, small-model ecosystem architecture, AI-assisted modernization, durable project handoff, source-governed research publishing, and .NET/SQL/TypeScript modernization.

Credibility summary

Michael's strongest contribution to ModelBreeder is practical engineering judgment. The site argues that adaptive AI should not be treated as a magical black box. It should be built like a versioned ecosystem with contracts, tests, evidence, release gates, source memory, and human review. His public portfolio shows the same pattern in conventional software: preserve behavior, expose seams, type the boundaries, measure the result, and keep the project understandable for the next reviewer.

pseudocode
FUNCTION build_credible_adaptive_system(concept, evidence, release_path)
    require public_claims_are_traceable(evidence)
    require interfaces_are_typed(concept)
    require descendants_have_lineage(concept)
    require tests_cover_behavior_before_promotion(release_path)
    require humans_can_review_and_contact_owner()

    return system_that_can_evolve_without_losing_accountability
END FUNCTION

Source trail

This page was synthesized from public pages on MikeKappel.com, mjk.tel, GitHub, NuGet, and Flickr. The crawl notes are preserved under /docs as long-term source memory and referenced from active .uai memory.

Source reports used for this guide

These reports are preserved verbatim in the site archive. The guide above is an editorial synthesis and may narrow, qualify, or reorganize claims from the source material.