Text so far
The best pet is a
Harris Oldroyd · Public Notebook
Harris Oldroyd
Local AI, systematic research, and the systems behind both.
A public research notebook built from the hardware up: running models locally, testing trading ideas against data, and building the software the work needs.
Start here
3 entry points01
Run your first model
Compare Ollama, llama.cpp, and vLLM, then run your first model with a single command.
02
Choose hardware
Evaluate GPUs, workstation cards, and unified memory across bandwidth, VRAM, and budget.
03
Understand inference
Step through tokenization, memory use, and KV cache mechanics with interactive visualizers.
What I work on
Three areas, one notebook.
Local AI is the deepest area right now; the research and software around it stay public alongside it.
Local AI
Run and understand models locally.
Hardware, quantization, runtimes, benchmarks, and interactive explanations for local inference.
Explore local ai →ResearchSystematic trading evidence.
Reproducible experiments on market regimes, filters, and execution — including the ideas that fail against data.
Explore research →SoftwareTools that make research repeatable.
Local-first projects for data, measurement, and research workflows.
Explore software →Local AI
The deepest area of the notebook.
Hardware, models, runtimes, and benchmarks for running AI locally — with a learning hub for going deeper.
Learning hub
Learn how local AI actually works.
Learning paths for fundamentals, hardware, models, and software, with interactive explainers for tokenization, attention, and the KV cache.
The featured guides below cover the decisions that matter most: choosing hardware, understanding quantization, and what a 3090-class build can actually run.
From prompt to next token
Tokens and IDs
TheID 464␠bestID 1266␠petID 4273␠isID 318␠aID 257
GPT-2 BPE tokens. ␠ marks a leading space included in the token.
Token source
Prompt IDs use the GPT-2 vocabulary (r50k_base). Continuations use a small example distribution.
Meaning and context
pet · 4273
[0.2, −0.7, …]Embeddings represent meaning. Attention mixes in useful context from the other tokens before the next-token scores are formed.
Possible next tokens
dog
cat
fish
Sampled result: dog
How sampling works
Temperature reshapes the distribution; top-p keeps the smallest cumulative set at or above p, then renormalizes it. Zero temperature chooses the highest-probability candidate. A fixed draw keeps this toy example reproducible.
One token is added
The best pet is a dog
Back revisits the previous generation's Pick step. Changing that choice starts a new continuation from there.
Featured
Guides worth starting with.
Selected hardware, fundamentals, benchmark, and model-analysis work rather than a chronological feed.
Hardware
I Built a 48 GB VRAM Local AI Workstation with Two RTX 3090s
A dual RTX 3090 local AI workstation with 48 GB of distributed VRAM, Threadripper Pro, WRX80, 128 GB ECC RAM and enough expansion room for larger multi-GPU experiments.
Read the guide →Buying guideBest GPUs and Devices for Local AI Inference
A comparison of GPUs, workstation cards, unified-memory devices, and CPU inference setups for local AI, focused on price, speed, memory capacity, software support, and practical tradeoffs.
Read the guide →FundamentalsLLM Quantization Without the Alphabet Soup
A beginner's guide to local LLM quantization: what bits per weight means, how popular GGUF formats compare, when to choose Q2 through Q8, and why Q4 is a strong starting point.
Read the guide →BenchmarkRTX 3090 for Local AI Inference
A practical look at the RTX 3090 for local AI workloads, covering specs, VRAM, bandwidth, TFLOPS, used pricing, power limiting, LLM performance, image generation, and why it remains one of the best-value GPUs for running open models at home.
Read the guide →All areas
Recent work across the notebook.
The latest from every area — Local AI, research, and software.
essay7 min read
Does Shorter HBM Mean Less AI Memory Demand?
A technical and economic guide to HBM stack height, capacity, bandwidth, packaging yield, qualified high-speed supply, and the competing interpretations of lower-layer AI memory.
Model analysis7 min read
GLM 5.3 Flash: My Experience Using It
A practical review of GLM 5.3 Flash covering its 320B-A18B architecture, API economics against DeepSeek V4 Flash and Luna, coding behavior, quantized memory footprint, and my plans to run it locally on an M5 Ultra Mac Studio.
Hardware5 min read
Apple M5 Ultra for Local AI
An overview of the Apple M5 Ultra Mac Studio for local large language models, covering its UltraFusion design, 1.2 TB/s memory bandwidth, expected prefill gains, decode expectations against NVIDIA Blackwell workstation cards, and how it compares to DGX Spark.
Model analysis10 min read
Qwen 3.8 27B: The Practical Sweet Spot for Local AI
An in-depth guide to Qwen3.8-27B for local inference. We examine its Artificial Analysis Intelligence Index standing, native multi-token prediction (MTP), weights versus KV cache quantization, parallel context slots, and firsthand daily driver observations on a dual-GPU workstation.
Elsewhere in the notebook
The wider research practice.
Quant research, software projects, and technical writing each have their own section.
Quant research
Research notes and experiments.
Methodology, testing, failures, and iteration on systematic strategies.
Browse quant research →Software projectsSystems, tools, and experiments.
Projects spanning quantitative research, automation, data engineering, and personal interests.
Browse software projects →Essays and notesFollowing interesting questions.
Long-form writing wherever curiosity leads: technology, markets, AI, and decision-making.
Browse essays and notes →