Context engineering at work
Build a source packet for an AI task: choose current evidence, label conflicts and test a synthetic BIM handover example in your browser.
Open guidePractical guides for learning, coding, research and everyday work — with source checks built into the workflow.
Choose a small task with a visible input and a result someone can review.
Novice and intermediate learning, programming workflows, or everyday office tasks — pick one lane first.
Check hardware with CanIRun.ai, compare local apps and try one offline task before buying upgrades.
Explore local AI options → Ideas into action Build your next versionSix visual shifts for AI innovation, a seven-day experiment and a prompt to rethink your work. Inspired by OBF attendee notes.
Explore the innovation field guide → Novice and intermediate Learn AI one task at a timeTwo clear tracks: start with free tools and safety basics, then move into prompting, paid plans and small automations.
Open learning path → AI with code AI for programming workflowsPick in-IDE assistants, chat fallbacks and local models for code review, tests and repo-aware edits — with developer news beside the stack.
Open programming desk → Daily work AI for everyday office tasksMeetings, writing, research and automation tools grouped by job type so you can test one workflow at a time.
Open daily work desk →These are original workflow diagrams, not product screenshots or performance claims.
Compare two document revisions and return a change table with the exact section that supports each finding.
Output: change table, source references and unresolved questionsAsk for transformations, formulas and totals to be shown so a reviewer can trace the result back to source rows.
Output: cleaned table, calculations and a visible checkGive an agent acceptance criteria, inspect the diff and run the relevant tests before accepting its handoff.
Output: reviewed diff, test result and remaining questionsEvery guide links to official vendor pages and keeps human review on anything contractual.
Build a source packet for an AI task: choose current evidence, label conflicts and test a synthetic BIM handover example in your browser.
Open guideMCP (Model Context Protocol) lets a compatible AI application use tools and context exposed by a server. Start with public documentation, confirm a real tool call, then connect a small test workspace.
Open guideChoose from nine publisher-maintained MCP servers by task, connection type and first test. Each entry links to the official setup documentation and carries a verification date.
Open guideAdd the public OpenAI documentation MCP to Codex, check the configured server and verify a source-backed tool call before connecting private accounts.
Open guideConnect GitHub's official remote MCP in VS Code, authorise your account and verify a read against one test repository before enabling changes.
Open guideConnect Notion's remote MCP to Claude Code and turn a synthetic page into an evidence-linked action table. Confirm access and source text before attempting a write.
Open guideDiagnose MCP problems in order: configuration, server startup, authentication, tool discovery and one small read. Capture the exact error before changing permissions.
Open guideCompare two synthetic specification excerpts, preserve uncertainty and produce an RFI draft with evidence. You can complete this exercise in an approved chat tool without MCP.
Open guideWhere AI helps BIM teams today: meeting summaries, spec search, RFI drafts and document QA — without replacing clash sign-off or contractual review.
Open guideHow to decide between a consumer chat subscription, an API budget or a team Copilot seat using your real weekly tasks and the live model value ranking on this site.
Open guideAn honest look at what the free tiers of ChatGPT, Gemini, Claude, Copilot, NotebookLM and Perplexity cover for everyday work and study, and when upgrading starts to matter.
Open guideChoose AI tools by job type: writing, coding, research, meetings, images and BIM workflows without buying every subscription.
Open guideA student guide to using AI for lecture summaries, exam preparation and course-material search without breaking academic integrity rules.
Open guideChoose between local AI agents and cloud agents for work, study and automation, with setup paths for Ollama, LM Studio and major cloud assistants.
Open guideA source-linked introduction to GPT-6 Astra for research, coding and document work, with a small test plan instead of a universal model ranking.
Open guidePick one coding agent for daily work — AI-native IDE, extension, terminal agent or desktop Codex — without subscribing to every developer tool at once.
Open guideChatGPT Work, Claude Cowork and Microsoft 365 Copilot agents — when computer-use agents are worth the risk, and when a chat draft is still the safer tool.
Open guideConcrete prompt patterns for meetings, reports, code review, client emails and BIM document tasks that reduce rework instead of creating more editing.
Open guideEvery link goes to the vendor's own site. Pricing type is noted; exact plan prices change too quickly to hard-code here.
Edits code in-repo with context instead of copying snippets into a separate chat window.
Official siteAutocomplete, agent mode in supported IDEs, and GitHub cloud agents for issue-to-PR workflows.
Official siteTerminal, IDE and Slack agent that reads your repo, runs commands and can prepare pull requests after permission prompts.
Official siteSoftware-development mode in the ChatGPT desktop app for repo-aware technical tasks separate from Chat and Work.
Official siteFast drafting for scripts, regex, SQL and one-off debugging when IDE integration is not required.
Official siteStrong long-document rewriting, structured summaries and careful tone control.
Official siteDrafts in Word, Outlook, Excel and Teams without leaving the Microsoft stack.
Official siteUseful when Docs, Gmail and Workspace files are already the daily environment.
Official siteSummarises meeting notes, wikis and project docs where Notion is the source of truth.
Official siteTurns meetings into searchable transcripts, summaries and action lists.
Official siteRecap, actions and follow-up drafts for Teams meetings already on Microsoft 365.
Official siteMeeting summaries and chat catch-up inside existing Zoom workflows.
Official siteWeb-grounded answers with citations for market, product and technical research.
Official siteQ&A over uploaded PDFs, specs, slide decks and project document packs.
Official siteFind open models, datasets and deployment options when you need more control than a chat app.
Official siteCommercial-safe image generation and edits inside Creative Cloud workflows.
Official siteFast concept visuals, mood boards and presentation imagery.
Official siteQuick social graphics, slide visuals and brand-safe layouts for non-designers.
Official siteVoiceovers, narration and dubbing without booking studio time for every draft.
Official siteRuns open-weight models on your own PC or server with a simple CLI and local API for private chat and scripting.
Official siteDesktop app for downloading, chatting with and serving local models without sending prompts to a cloud API.
Official siteBrowser UI for local models with chat history, document upload and team-friendly self-hosting on top of Ollama or compatible backends.
Official siteOffline-first desktop chat for local models when you want a simple app without running a server stack.
Official siteMulti-step office agent in ChatGPT for finished docs, sheets and slides with connected apps and supervised actions.
Official siteDesktop agent for folder-based tasks, browser steps and multi-step deliverables with permission prompts.
Official siteAgents inside Copilot that act on Excel, mail, records and workflows with enterprise governance options.
Official siteGems and Workspace-connected chat when your files, mail and docs already live in Google's cloud stack.
Official siteResearch agents with cited web answers when the task needs live sources rather than only private files.
Official siteConnects apps with no-code automations for repetitive admin and handoffs.
Official siteFlexible workflow automation when you need custom logic or self-hosting.
Official siteHelp inside Autodesk workflows for document search, drafting support and product guidance.
Official siteSpec comparisons, transmittals, coordination emails and Excel-based quantity review support.
Official siteProject briefings, method statements and onboarding packs from uploaded PDFs and reports.
Official siteChoose current evidence, spot an obsolete approval and build a checkable source packet.
Try the context-engineering guide and exercise →#1 · 100 price/capacity index · 17.2M tokens/$
OpenRouter page#2 · 55.4 price/capacity index · 10.0M tokens/$
OpenRouter page#3 · 51.2 price/capacity index · 9.8M tokens/$
OpenRouter page#4 · 45 price/capacity index · 7.7M tokens/$
OpenRouter pageA conversation with Sam Nickless, Partner and Chief Executive Officer at Gilbert + Tobin.
Read OpenAI NewsMore than 20,000 veterans and transitioning service members have claimed a free year of ChatGPT Plus. See how they use it to prepare for civilian careers.
Read OpenAI NewsOpen the original announcement for confirmed product details, availability and pricing.
Read Hugging Face BlogRelease: llm-gemini 0.36 Gemini 4 Carbon was announced two weeks ago for "a subset of cyber defenders", but today it's available to everyone else . This is Google's best ever Gemini, priced at $2/million input and $10/million output at launch (that's GPT-6.1 Sol / Claude Sonnet 5.5 prices) but scheduled to double in price to...
Read Simon Willison