Build a building-code research checklist with AI, then verify every section
Use a code search platform plus a general AI assistant to pull applicable IBC and local amendment sections, then verify each citation.
The practical AEC workflow library
Find a practical way through your next design, engineering, or construction task. See what runs, what needs you, and where to start.
Free guides with steps, starter prompts, and honest limits. Choose a problem, then put the workflow to work.
Use a code search platform plus a general AI assistant to pull applicable IBC and local amendment sections, then verify each citation.
AI visualization plugins restyle a SketchUp, Revit or Rhino view into photoreal options in minutes for early design conversations.
Feasibility software generates site plans and yield options from zoning and parking inputs so you can compare scenarios quickly.
Cloud space-planning tools import site and Revit data and measure automatically as you edit.
BIM automation products aim to generate documentation using a firm's own standards.
Calculation management tools put calcs in a shared, versioned workspace; an AI assistant can draft the narrative and check units.
Design data platforms connect Revit, Rhino and IFC, validate each model version, and let you ask questions of live model data.
The Model Context Protocol (MCP) lets AI assistants connect to tools. Open-source Revit MCP servers expose Revit actions to an agent.
Design automation software for critical infrastructure produces preliminary designs and options early in a project.
Reality capture platforms turn drone flights into maps, models and measurements.
Spatial AI tools aim to automate building system design such as electrical routing.
Building performance software helps teams test energy, daylight and carbon early in design.
LCA software calculates embodied carbon and supports EPD work.
AI document review tools built for construction flag risk in contracts and project documents.
Construction AI platforms read project documents so teams can ask questions and get sourced answers.
AI scheduling tools generate and compare schedule scenarios and forecast risk.
Project controls software analyzes schedule updates for delays and quality issues.
AI takeoff tools detect and measure areas, counts and lengths on plans.
AI estimating assistants for residential builders create estimates and takeoffs and manage budgets.
A general AI assistant plus document search can draft submittal cover letters, RFIs and spec summaries for steel and precast shops.
360 camera capture and computer vision platforms track progress against plan and schedule.
AI plan review tools check plans against codes such as ICC, NFPA, ADA and local amendments.
Computer vision safety platforms detect hazards and compliance issues from site cameras.
Jobsite management apps track tasks and drawings; an AI assistant can turn notes into tasks.
BIM collaboration platforms combine models and sheets, track issues and push them back to authoring tools.
Connection design software checks steel connections and concrete details against AISC, Eurocode and other standards.
Browser-based structural analysis and design software covers frames, beams and member design.
Calculation platforms provide reviewable, traceable member calcs with standards built in.
Model-based construction software creates detailed steel models, drawings and fabrication data.
Civil-focused AI takeoff tools measure earthworks, drainage and concrete and flag discrepancies.
Pre-construction AI reviews drawings, specs and contracts to surface risks and scope gaps.
Precon estimating platforms let teams build and update conceptual estimates collaboratively.
Bid management platforms connect GCs with subcontractor networks and track bid invitations and responses.
Project lead databases and AI pursuit tools surface upcoming projects and help prioritize them.
Material management software tracks requests, orders and deliveries for trade contractors.
Workforce planning software helps contractors allocate people across projects.
Field reporting apps capture daily reports, time and safety; an AI assistant can clean up dictated notes.
Construction defect and inspection apps tie issues to plans with photos and assignees.
3D capture platforms create immersive tours, measurements and schematic floor plans.
Reality intelligence platforms compare site capture to plans and schedules to find deviations.
Autonomous drones support critical infrastructure inspection and mapping.
Zoning intelligence tools find development sites and model returns.
Capital project management software tracks budgets, schedules and risks across projects.
Project information management platforms centralize files, emails and conversations.
Workplace operations apps run digital checklists, inspections and corrective actions.
Write a step-by-step extraction skill, run an AI model over the shop drawing set, and verify every cell with cross-checks such as plate weight from geometry vs. the bill of materials. In our own trial on 21 connections (273 values), the best model run matched 268 of 273 values (98%) against the verified answers; verification rules made the biggest difference.
Generate an unlocked spreadsheet from a script, cite spec section numbers next to every formula, and test it against an independent hand calc so the workbook can be reviewed instead of trusted as a black box.
A scheduled job checks official publisher pages (ICC, ASCE/SEI, AISC, ACI, AWC, AWS, state code agency) for new errata, supplements and adoption notices, and an AI drafts a weekly digest that separates fact from implication.
Many cities publish issued permits through open data APIs. A short script pulls the last 7 days, filters by work type and job cost, and writes a CSV you can sort for bidding or outreach.
Join the submittal log to schedule activities, subtract lead time and review time from the need-on-site date, and email the team a list of submittals that must move in the next two weeks.
A Dynamo graph or pyRevit script reads viewport positions on each sheet, groups them into rows and columns, and assigns detail numbers in your grid convention without the 'Detail Number is already in use' error.
Use an AI coding assistant to rebuild a PDF calculation package (for example a connection plate redesign) as a spreadsheet: inputs in highlighted cells, then one check per block (gross yielding, rupture, combined axial and bending, buckling, block shear), each matched to the PDF value.
With MCP connectors to both Revit and Unreal Engine, an agent reads the Revit 3D view as the source of truth and sets up the Unreal level and sequencer for stills and fly-through video.
An agent connected to Blender through an MCP add-on builds simple massing (floors, collections), sets cameras and renders short development marketing clips you can iterate on in plain language.
When an access opening is cut into a tank shell after design and the shell deforms near it, use AI research to gather reinforcement approaches, then plan a shell finite element comparison (no opening, as-built opening, proposed reinforcement) checked against hand hoop-stress calcs.
A standing rule in the agent's instruction file (for example AGENTS.md) makes every engineering task log record tokens and cost, so you learn what AI help actually costs per task type.
Keep panel schedules and the load calc in one Excel workbook, apply demand factors from a load-type column, and use AutoCAD Data Link so the drawing table updates when the spreadsheet changes.
A shared sheet stores each sub's required documents and expiration dates, flags anything within 30 days, and acts as a hard gate: nothing expired gets scheduled or paid. A weekly reminder email does the chasing.
One workbook holds the estimate summary, detail, qualifications and a leveling tab per trade. A shared cost-code column links them, so the number you carry on each leveling tab flows into the estimate with SUMIFS instead of retyping.
Upload a borehole log PDF to Claude or ChatGPT, have it extract depths, strata descriptions, SPT or RQD values and groundwater notes into a fixed column layout, then spot-check every row against the PDF before you import anything into gINT, OpenGround or your own workbook.
One shared sheet lists every vendor and pay period with conditional-sent, payment-released, unconditional-due and unconditional-received columns, so payment cannot move when the trail is incomplete. Copilot or Gemini builds the sheet and drafts the chase emails.
A per-project workbook with tabs for contract value and changes, pay applications, commitments, schedule milestones and a one-page dashboard. Copilot in Excel builds the structure and formulas; you keep the numbers honest against the GC pay apps.
During construction, RFIs pile up across email and folders. Have Copilot in Excel or Gemini in Sheets build a log with number, subject, date received, due date, consultant, date sent, date returned, status, priority, working days elapsed and an overdue flag with conditional formatting, then draft a short OAC blurb from open rows only.
Upload division or section PDFs to Claude or ChatGPT and have it list required submittals (section, description, type: shop drawing, product data or sample) into a fixed column list, marking unclear rows UNVERIFIED. Then have Copilot build an Excel register with required-on-site, lead time, submit-by formula, review due, status and overdue flags. Distinct from back-calculating dates when a register already exists.
When dozens of similarly structured field workbooks need one master for review or closeout, Claude, ChatGPT or Copilot can propose a Python or Power Query approach, or map columns from uploaded sample sheets into one schema. Unmapped columns stay UNVERIFIED; field readings are never invented.
Banks and sureties read the work-in-progress schedule closely, yet many companies run it on hidden columns and nested IFs that one person understands. Have Copilot in Excel or Claude rebuild it as three tabs: raw job data, transparent cost-to-cost calculations (percent complete, earned revenue, over/under billing, projected gross profit, loss flags) and a clean report, then tie the totals to the general ledger yourself.
For co-ownership or condominium plans, each lot's private area is often measured by hand from closed polylines, minus shafts and columns, with terraces and gardens kept separate. Have Claude or ChatGPT write a short, commented AutoLISP routine (or a Python script using ezdxf on a DXF export) that reads closed polylines on layers you tag, subtracts voids inside each lot and exports one row per lot to Excel. You check the drafting, the tagging and the local measurement rules.
At turnover, or when catching up on missing data, upload equipment O&M manuals one at a time to Claude, ChatGPT or Copilot and have it extract asset details and the manufacturer's maintenance tasks into a fixed table with the manual page on every row. Gaps are marked NOT IN MANUAL for the manufacturer, not filled in. The same basis column then helps you review old PM tasks nobody can justify anymore.
Field crews need more detail than the monthly CPM schedule carries, so most superintendents keep a separate three-week look-ahead in Excel. Export the activities that fall in the window from P6 or Microsoft Project, have Copilot in Excel or Claude draft a breakdown of each one by trade, area and day for you to edit, and add flags for two trades in the same area on the same day. The master activity ID stays on every row, so the look-ahead still ties back to the schedule.
Large equipment submittals such as air handlers, chillers, pumps and switchboards take the longest to review. Upload the product data with your equipment schedule row and the spec items you care about to Claude or ChatGPT, and ask for a table that compares each design value to what was submitted, with page references and a MATCH, DIFFERENT or NOT SHOWN status. You review from that checklist; the comments and the review action stay with you.
Near-miss and observation reports usually arrive as short free-text notes, so repeat patterns get missed. Paste a month of reports, names removed, into Copilot in Excel, Claude or ChatGPT and have it tag each one with your own fixed hazard categories plus trade, task and area, marking anything unclear as UNCLEAR. A pivot then shows what keeps repeating, and the model drafts toolbox talk outlines built only from those reports for you to edit.
Draw reviews come down to the same checks every month: does each requested dollar sit on an approved budget line with an invoice behind it, does the remaining balance still cover the work left, and is retainage held the way the loan documents say. Paste the borrower's request, the approved loan budget, your own draw history and the invoice list into Copilot in Excel or Claude and have it build a line-by-line reconciliation with an exception list. The inspector's site report, title and lien checks, and the funding decision stay with the lender's team.
Extra work gets lost when daily T&M tags sit in a truck until month end. Photograph each signed tag, have Claude or ChatGPT transcribe it into a fixed table (crew by trade, hours, equipment, materials, description, reference number), marking anything it cannot read as UNREADABLE, then have Copilot in Excel price the rows with the labor rates, equipment rates and markups you enter from your subcontract and roll them into a change order request log. The tags document what happened; entitlement and price still have to be agreed.
Many analysis programs report plate results as separate Mx, My and Mxy, so engineers export them to Excel, combine them into design moments by hand and then search the model for the governing plates. Have Claude, ChatGPT or Codex write a short, commented Python script that applies the design-moment equations you paste from your own reference, assigns plates to the design strips you define by coordinates, and lists the governing plate and load combination per strip. You choose the method, the strips and the sign convention, and you hand-check a few plates.
Try fewer words, select “All” for one filter, or .
Choose the right workflow for the job.