AI Software & Tools
AI coding assistants, tools for comparing language models, and AI writing tools: what they do, who they suit, and what was verified.
Practical write-ups from Rajib Roy, an AI developer since 2015: AI tools, the hardware that runs them, and the cloud services behind them. Code is run before it is published, and sources are named.
AI coding assistants, tools for comparing language models, and AI writing tools: what they do, who they suit, and what was verified.
PCs, laptops and GPUs for running AI, including local LLMs: what fits in memory, what to buy and what to skip.
GPU cloud, LLM APIs, hosting and vector search: what they cost, how they fail, and how to keep them under control.
Tutorials for building AI features: evaluation sets, prompt versioning, prompt-injection defenses, and streaming, with code that has been run.
Documentation, READMEs, and small developer utilities that are not AI-specific.
A tested Node.js server and fetch-based SSE reader that streams tokens, stops paying for abandoned requests, and plugs into a small React hook.
A README template that answers what, why and how in the order readers need, what GitHub does with your README, and a tested Python linter that flags missing sections, untagged code blocks and broken relative links.
A small Node.js pattern for versioned prompt files with strict variables and a content hash in every log line, so you can always answer which prompt produced which output.
Stop retyping the same instructions. Build a small library of prompt snippets with /slash expansion in Node.js, plus how chat tools like Shortcut Chat handle reusable shortcuts.
A practical priority list for web developers adding LLM features: limit what the model can do, validate tool calls in code, and require confirmation for side effects. Includes a tested tool-call gate.
When pgvector or even a plain in-memory scan is enough for semantic search, with a runnable demo of the filtered-search trap that returns too few rows and the pgvector 0.8 fix.
Build a 30-case evaluation set, a 40-line runner and a CI check so prompt changes can't quietly break your LLM feature. Includes a runnable Node.js example with a baseline file.
A tested Node.js module that adds a per-user daily token budget and a TTL response cache around any LLM call, plus the habits that keep an AI feature's bill predictable.
A short, tested Node.js module that adds a hard timeout, backoff with jitter, Retry-After handling and schema-validated JSON to any chat-style LLM API call.
Build a PDF invoice with PDFKit and Node.js: integer-cent money, correct tax rounding, multi-page tables with repeated headers and page numbers, and tests that read the generated PDF back and check the numbers.
Organise developer docs around four reader needs (tutorials, how-to guides, reference, explanation), write examples that run, keep a useful changelog, and check structure and links automatically with a tested audit script.
A practical, blind-scoring method for choosing between AI models for a real feature, with a tested Node.js scorer, plus where side-by-side chat tools like Shortcut Chat help and where they stop.
Where AI-generated Solidity needs human review: tx.origin authorisation, ignored ERC-20 return values, and unchecked upgrade and access assumptions, with failing tests you can run.