Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: GPT
Training a Small GPT on Indian Folk Tales: Tokenizer, Mask and Loss
A from-scratch 9M-parameter GPT: why GPT-2's tokenizer fails on folk-tale words, how BPE and the causal mask work, and how one batch yields 4,096 next-token losses.
16 min readWhat GPT Really Stands For and Why AI Engineering Is Still Engineering
Unpack Generative Pre-trained Transformer, attention, fit and predict, and see why good AI work still depends on ordinary engineering judgment and better questions.
4 min readPractical notes: What Does GPT-5.5 Mean for AI Agents and Customer Support
Operable walkthrough of Practical notes: What Does GPT-5.5 Mean for AI Agents and Customer Support: contracts, checks, and drop-in code slots for teams shipping this pattern.
7 min readPractical notes: GPT-5.6 Sol at 750 Tokens/Second: What Cerebras-Speed
Operable walkthrough of Practical notes: GPT-5.6 Sol at 750 Tokens/Second: What Cerebras-Speed: contracts, checks, and drop-in code slots for teams shipping this pattern.
10 min readFrom GPT-1 to Reasoning Models: What Each Generation Changed for Developers
Trace the GPT family from 2018 pre-training to reasoning models, see which ideas each generation added, and call GPT-4o vision and Structured Outputs from Python.
10 min readFugu Ultra: How an AI Orchestrator Model Challenges GPT and Claude
Explains how Sakana AI's Fugu Ultra v2 routes tasks across specialist models instead of one giant LLM, and how it stacks up on benchmarks, pricing, and transparency.
10 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.