AI-Augmented Knowledge Work
AI Tools That Reduce Context-Switching for Deep Work
Protecting focus time and memory matters more than automating the most tasks.
Solange Okafor
· 8 min read
The Locally Hosted AI
AI-Augmented Knowledge Work
Local AI Assistants for Managing Multiple Projects Without Losing Thread
Local AI assistants keep project context persistent so you skip the daily reload.
Marcus Llewellyn9 min read
AI-Augmented Knowledge Work
Note-Taking and AI Integration Tools That Keep Reasoning Local
Local AI note tools let you choose where your data and processing actually happen.
Priya Venkataraman8 min read
AI Security
Data Leakage Patterns in AI Assistant Plugins and Extensions
Three attack patterns exploit the gap between permissions users grant and behaviors they understand.
Tobias Gerhart9 min read
Open-Weight Models
Building a Self-Hosted Memory Layer Around Open-Weight Models for Lasting Work Context
Combining memory, retrieval, and context windows preserves work across sessions.
Solange Okafor9 min read
Model Benchmarks
First Token Latency vs Throughput Trade-Offs in Local Inference
Optimizing for speed requires choosing between fast first tokens or high output volume.
Tobias Gerhart8 min read
Model Benchmarks
Tokens per Second Benchmarks Across Consumer GPUs
Memory bandwidth, not raw compute power, determines how fast GPUs generate tokens.
Tobias Gerhart9 min read
Model Benchmarks
Reasoning Model Benchmarks for Local Deployment
Benchmarking reasoning models requires measuring hardware constraints, not just accuracy scores.
Solange Okafor8 min read
Model Benchmarks
Summarization Quality Benchmarks for Local AI Models
Separating groundedness, completeness, and factual accuracy reveals what ROUGE scores miss.
Tobias Gerhart8 min read
Model Benchmarks
Coding Task Performance of Small Local Models
Small local models excel at routine coding tasks but struggle with complex multi-step work.
Marcus Llewellyn7 min read
Model Benchmarks
Instruction-Following Accuracy of Local Models on Structured Tasks
Local models fail structured tasks through logic, timing, and uncertainty—not just format.
Priya Venkataraman8 min read
Data Sovereignty
Auditing Network Traffic From Local AI Tools
Discover which network channels leak data from local AI tools and how to audit them.
Solange Okafor9 min read
