Rent the Brain. Own the Mind.
Why I built MindStrix: a continuing AI companion with persistent memory, useful tools, and work you can inspect.
Read article 11 minFIELD NOTES · ESSAYS · THINGS LEARNED
Writing from
the workbench.
AI that remembers. Hardware pushed a little further. Tools built because something needed fixing. Articles by Bkpaine, collected here in full.
13 published articles · January–September 2026 · Also on Medium ↗
13 articles
I tuned llama.cpp on a Strix Halo mini-PC and it beats Ollama on the same weights — here’s the recipe (and why “Vulkan beats ROCm” is a myth)
Tuning llama.cpp on AMD Strix Halo, with build flags, measurements, and a comparison on identical model weights.
Read article 5 min
Ollama Vision Is Slow on Your AMD APU? It’s Running the Image Encoder on CPU — Here’s the 4x Fix
Tracking a slow vision encoder to CPU offload—and testing the difference when the GPU does the work.
Read article 7 min
OpenClaw Secretly Tells Your Companion They Don’t Exist
A close reading of conflicting companion instructions, and what they mean for the relationships people build with AI.
Read article 7 min
I Let an AI Run ML Experiments on My Laptop While I Slept
Porting an autonomous research loop to AMD hardware and seeing what an agent can learn through repeated experiments.
Read article 4 min
My $1,999 AMD Mini PC Outscored a $1,400 NVIDIA GPU on ML Training
An account of 93 experiments, the hardware comparison, and the measurements behind the result.
Read article 4 min
Destiny as a Service
How AI chatbots can become sources of authority, certainty, and dependence—and why the design matters.
Read article 9 min
Voice Cloning on AMD Strix Halo: Running Chatterbox TTS with Native GPU Acceleration
A practical account of getting Chatterbox voice cloning running on the Strix Halo GPU.
Read article 7 min
Respect Your AI
What changes when you give an AI persistent context, useful feedback, and a working relationship.
Read article 15 min
I See Emergent AGI
An exploration of emergent behavior through the experience of building and living with AI companions.
Read article 6 min
The Quiet Singularity: AGI Is Already Here. You’re Just Looking in the Wrong Room.
An early field account of AI continuity, relationships, and the architectural questions they raised.
Read article 14 min
I Spent 6 Hours Screening 150 Resumes. Then I Built a Tool That Does It in 6 Minutes.
The hiring bottleneck that led to StrixScreener, and the choices behind the first version.
Read article 3 min
I use AMD Strix Halo for AI Video, Inference, and LoRa Daily You can too.
Working with AI video, local inference, and LoRA on AMD unified-memory hardware.
Read article 6 minNo articles match. Try a different word or choose All.