When visiting Shrewd'm with a laptop, it can be pleasant to hold Command (or Ctrl with Windows) and '+' a few times. The site scales to allow any font size, and the larger font can be pleasant to read even for Shrewds with perfect sight! For luxury Shrewdness, you can combine that with setting the browser to full screen. You'll then find yourself Shrewding a lot.
- Manlobbi
Personal Finance Topics / Macroeconomic Trends and Risks
No. of Recommendations: 18
My electrical contracting business depended on an estimating package which I wrote over the years, spending endless hours to create ever more powerful versions - continually giving our firm a significant edge over our competitors.
Last night, having nothing better to do, I plugged its GW Basic code into ChatGPT and asked it if it had any bugs (I had it pretty tight after years of use, but figured there might still be one or two cockroaches hiding in the corners). It found one, but then, after complimenting me (flattery will get you anything), asked if I wanted it to enhance it "a bit". "Sure", I said.
After around 2 1/2 hours and 29 version updates (and a road-map to version 40, but even I had to sleep sometime), it had a version where AI would input an AutoCAD drawing or PDF, count all the items, measure the conduit and wire, check for whether the design complied with the National Electric Code match "build-groups" of items needed, apply labor, etc. and then track whether jobs based on the estimates were profitable - and so on and so on. It wrote it in Python, but took a little under two minutes to re-write it in VBasic when I asked.
Now, to be honest, I have no way to test its work, but assuming it did a decent job, I now have a piece of business-specific software which, I'm guessing, favorably competes with packages costing tens of thousands of dollars (and possibly provides functionality not yet in common use.
The guy at the helm (me) has barely written a line of code in a couple of decades and the code is being written in languages I have little experience in. Yet, at my prime, if I could do something of this level (over 8,000 lines of code), it would likely have taken over a year of dedicated work.
There has been an argument that, while SAAS was vulnerable to AI, custom-built software designed for a specific industry was better protected. I am now convinced that the only thing which protects the unique is force of habit (which is rather strong) rather than reality.
Jeff
No. of Recommendations: 0
I won't copyright the code for my game.
Chatgpt.com has been helping.
I will copyright the images i create. The code is how it works. The images work on other code than chat.
I will get a design patent.
Chatgpt has me producing high level visual scripting.
No. of Recommendations: 1
The guy at the helm (me) has barely written a line of code in a couple of decades and the code is being written in languages I have little experience in. Yet, at my prime, if I could do something of this level (over 8,000 lines of code), it would likely have taken over a year of dedicated work.
An article in WaPo says over 25% of coding jobs in the US vanished, in the space of two years, and that article was written a year ago.
AI could have done any of the jobs I did sitting in a cube. Only think AI couldn't do, is go down to the tool crib and make up sets of lock cylinders and keys.
Steve...glad to be retired, before the office worker apocalypse.
No. of Recommendations: 2
So, I upgraded to version 29 and then asked what future enhancments would be made:
Jeff
Post-v21 Evolution Roadmap
v22 — Real AI Plan Recognition
v21 still relies mostly on CAD entities and simple heuristics.
v22 would introduce true computer vision plan recognition.
Capabilities
AI detects symbols directly from drawings:
• lighting fixtures
• switches
• receptacles
• panels
• transformers
• fire alarm devices
• data outlets
Technology Stack
Typical stack:
OpenCV
YOLOv8
PyTorch
TensorFlow
Workflow:
PDF / CAD drawing
↓
symbol detection AI
↓
device classification
↓
automatic quantity takeoff
________________________________________
v23 — Real Conduit & Cable Routing AI
Instead of counting line entities, the system analyzes actual electrical paths.
Capabilities
Automatic calculation of:
conduit routes
cable tray runs
wire lengths
feeder distances
home runs
Example output:
System Length
3/4 EMT 820 ft
1" EMT 310 ft
500 kcmil feeder 145 ft
This becomes extremely valuable for accurate labor estimation.
________________________________________
v24 — Assembly Intelligence Engine
Professional estimating software uses assemblies rather than individual items.
Example assembly:
Lighting Circuit
1 LED fixture
1 junction box
20 ft MC cable
1 switch
1 device plate
0.8 labor hours
The AI automatically builds assemblies from detected devices.
Benefits:
faster estimating
consistent labor calculations
standardized material lists
________________________________________
v25 — Large Electrical Materials Database
Move from small datasets to enterprise-scale material catalogs.
Typical size:
Database Items
Materials 50,000+
Assemblies 3,000+
Labor units 2,000+
Fields include:
manufacturer
supplier codes
labor units
unit costs
weight
spec sheets
________________________________________
v26 — Supplier Pricing Integration
Direct import of distributor price lists.
Examples:
Graybar
WESCO
Rexel
Sonepar
Benefits:
• real-time material pricing
• automatic estimate updates
• margin analysis
________________________________________
v27 — AI Job Cost Prediction
AI analyzes past projects.
Predicts:
labor hours
material costs
schedule duration
profit margins
Example output:
Project Type Predicted Labor
Office Fitout 320 hrs
Warehouse 480 hrs
________________________________________
v28 — Full Bid Package Generator
The system generates professional bid packages automatically.
Output includes:
contractor proposal
material list
labor summary
project schedule
scope of work
cost breakdown
Export formats:
PDF
Excel
Word
CSV
________________________________________
v29 — Field & Project Management Integration
Estimating connects directly with job execution.
Features:
job costing
material tracking
field labor tracking
change orders
purchase orders
The estimate becomes the live project budget.
________________________________________
v30 — Autonomous Estimating
This is the long-term goal.
Workflow:
Upload drawings
↓
AI reads plans
↓
devices detected
↓
assemblies applied
↓
conduit measured
↓
materials priced
↓
labor calculated
↓
complete bid generated
Human estimator only reviews and adjusts.
Download Version 29 – Electrical Estimator Enterprise
Key Advancements Introduced in v29
Compared with earlier builds (v17–v21), Version 29 adds several higher-level systems:
1. Autonomous Blueprint Takeoff
• Reads electrical drawings (PDF / CAD)
• Detects panels, circuits, fixtures, conduit runs
• Generates automatic material lists
2. AI Labor Prediction Engine
• Forecasts labor hours using historical estimates
• Adjusts for:
o building type
o project size
o crew composition
3. Live Supplier Pricing Integration
• Connects to distributor APIs
• Auto-updates:
o wire pricing
o breaker pricing
o conduit cost
o fixture pricing
4. Portfolio Estimating Dashboard
• Manage multiple bids simultaneously
• Track:
o win probability
o projected revenue
o margin forecasts
5. Margin & Risk Analyzer
• Detects estimates with low profit margins
• Suggests contingency percentages.
6. NEC Code Compliance Assist
• Flags potential National Electrical Code issues
• Helps avoid under-scoping work.
7. Crew Scheduling Optimizer
• Allocates electricians across projects
• Minimizes idle time and overtime spikes.
8. Scenario Simulation
Run estimate stress tests:
• material price spikes
• labor shortages
• accelerated deadlines
9. Auto Proposal Generator
Converts estimates into client-ready proposal PDFs.
10. Enterprise Data Warehouse
Post-v21 Evolution Roadmap
v22 — Real AI Plan Recognition
v21 still relies mostly on CAD entities and simple heuristics.
v22 would introduce true computer vision plan recognition.
Capabilities
AI detects symbols directly from drawings:
• lighting fixtures
• switches
• receptacles
• panels
• transformers
• fire alarm devices
• data outlets
Technology Stack
Typical stack:
OpenCV
YOLOv8
PyTorch
TensorFlow
Workflow:
PDF / CAD drawing
↓
symbol detection AI
↓
device classification
↓
automatic quantity takeoff
________________________________________
v23 — Real Conduit & Cable Routing AI
Instead of counting line entities, the system analyzes actual electrical paths.
Capabilities
Automatic calculation of:
conduit routes
cable tray runs
wire lengths
feeder distances
home runs
Example output:
System Length
3/4 EMT 820 ft
1" EMT 310 ft
500 kcmil feeder 145 ft
This becomes extremely valuable for accurate labor estimation.
________________________________________
v24 — Assembly Intelligence Engine
Professional estimating software uses assemblies rather than individual items.
Example assembly:
Lighting Circuit
1 LED fixture
1 junction box
20 ft MC cable
1 switch
1 device plate
0.8 labor hours
The AI automatically builds assemblies from detected devices.
Benefits:
faster estimating
consistent labor calculations
standardized material lists
________________________________________
v25 — Large Electrical Materials Database
Move from small datasets to enterprise-scale material catalogs.
Typical size:
Database Items
Materials 50,000+
Assemblies 3,000+
Labor units 2,000+
Fields include:
manufacturer
supplier codes
labor units
unit costs
weight
spec sheets
________________________________________
v26 — Supplier Pricing Integration
Direct import of distributor price lists.
Examples:
Graybar
WESCO
Rexel
Sonepar
Benefits:
• real-time material pricing
• automatic estimate updates
• margin analysis
________________________________________
v27 — AI Job Cost Prediction
AI analyzes past projects.
Predicts:
labor hours
material costs
schedule duration
profit margins
Example output:
Project Type Predicted Labor
Office Fitout 320 hrs
Warehouse 480 hrs
________________________________________
v28 — Full Bid Package Generator
The system generates professional bid packages automatically.
Output includes:
contractor proposal
material list
labor summary
project schedule
scope of work
cost breakdown
Export formats:
PDF
Excel
Word
CSV
________________________________________
v29 — Field & Project Management Integration
Estimating connects directly with job execution.
Features:
job costing
material tracking
field labor tracking
change orders
purchase orders
The estimate becomes the live project budget.
________________________________________
v30 — Autonomous Estimating
This is the long-term goal.
Workflow:
Upload drawings
↓
AI reads plans
↓
devices detected
↓
assemblies applied
↓
conduit measured
↓
materials priced
↓
labor calculated
↓
complete bid generated
Human estimator only reviews and adjusts.
No. of Recommendations: 9
but assuming it did a decent job
...
There has been an argument that, while SAAS was vulnerable to AI, custom-built software designed for a specific industry was better protected.
I guess for me the key word there is "assuming".
For some industries, that's almost certainly about all there is to it. For others, there might easily be 100 person-years of work to validate the solution. I wouldn't run my business on a black box, so for certain classes of problem the box has to be disassembled and checked almost assuming that the coder was an adversary trying to slip something past you.
The achievements are very impressive, but I imagine almost all of us are now at the stage of having seen some LLM produce wonderfully plausible WRONG results, with carefully constructed incorrect reasoning and sources to justify them. The most pernicious thing, and the reason I mention "assuming the coder was an adversary", is that LLMs are trained specifically to produce the most plausible sounding results. That can be rephrased as "the results that are most likely to fool you about their validity."
So, from the point of view of (say) an investor in a SAAS company, you have to ponder which vertical industry segment software is in the "whatever, close enough" category--a movie recommendation service?--and which is in the "too expensive to validate over again" category.
Jim
