oss-model-playbook
Enterprise playbook for running open-weight LLMs on on-prem Kubernetes, AKS, and Azure AI Foundry — model selection, capacity planning, gateway, security, and TCO.
Active public repositories, fetched at build time from github.com/heisenberg-alt.
Assess, modify and implement GitHub Enterprise Cloud best practices — CLI + web wizard
Config files for my GitHub profile.
No description yet.
Composed-defense benchmark for prompt injection: grades entire defense stacks (model x detectors x defenses x capability policy) on ASR, FPR, latency, and cost. Open-weight-first for reproducibility.
No description yet.
Enterprise playbook for open-weight LLMs — on-prem K8s, AKS, and Azure AI Foundry. Interactive model picker, team cookbook (PDF), capacity planning, gateway, security and TCO. Live: https://heisenberg-alt.github.io/oss-model-playbook/
Longer write-ups — context, decisions, what I'd do differently.
Enterprise playbook for running open-weight LLMs on on-prem Kubernetes, AKS, and Azure AI Foundry — model selection, capacity planning, gateway, security, and TCO.
Composed-defense benchmark for prompt injection that grades entire defense stacks on attack success rate, false positives, latency, and cost.
Staff-level deep dive on embeddings, ANN, hybrid retrieval, agent memory, evaluation, and production ops for RAG and AI agents.
A token-saver bootstrap for repos that drops focused Copilot instruction files and VS Code defaults.
Estimator and explainer for GitHub Copilot usage-based billing after the June 2026 pricing transition.
The reference codebase + scripts I use when running GitHub Copilot Agent Mode workshops on real-world code migration.
Migration helper that translates GitLab CI pipelines into GitHub Actions workflows for faster platform consolidation.
Microsoft × ABB hackathon — building enterprise AI use cases on Azure that actually leave the demo phase.