Example projectWeb appAICustom LLM
Log count and diameter model
Hypothetical. Not a client, not a result.
// The problem
A mill checks loads against the scale by eye at the gate, and disputes over piece counts and log sizes are settled with nobody's evidence.
// What we would build
A custom vision model, fine-tuned on the mill's own photos of truck loads, that counts logs and estimates end diameters from a camera at the gate, running on a GPU server at the mill so it keeps working when the internet does not.
// What is in it
- A camera at the scale or gate, triggered as each load stops
- Piece counts and diameter estimates for each load, with the photo
- A model fine-tuned on your loads, measured against hand counts first
- Inference on a GPU server at the mill, with no photos sent away
- Differences from the scale ticket flagged for the scaler
// Stack
- PyTorch
- ONNX Runtime
- Python
- Next.js
- Lenovo GPU server at the mill
// estimate
- Build
- ≈ US$153,000 to US$233,000, delivered within 36 weeksCAD 218,500 to 332,400
- Hosting
- ≈ US$176 a monthCAD 250 a month
- Support
- ≈ US$2,500 a monthCAD 3,565 a month
- Hardware
- ≈ US$47,800 to US$197,000CAD 68,000 to 280,800
Prices in your currency are estimates from today's Bank of Canada rate. All invoicing is in CAD or USD.
- AI route
- Custom LLM
- Model running cost
- ≈ US$176 a monthCAD 250 a month
Timeline by milestone
- Discovery
- 2.9 to 3.4 weeks
- Data and evaluation plan
- 0.4 weeks
- Design approved
- 2.4 to 3.9 weeks
- Core features
- 1.9 to 3.4 weeks
- Full build
- 1.4 to 2.4 weeks
- Training
- 8.9 to 15.4 weeks
- Testing and fixes
- 0.4 weeks
- Evaluation and red-teaming
- 3.9 to 4.4 weeks
- Launch
- 0.9 to 1.4 weeks
- Production
- 0.9 to 1.4 weeks
Outside our hands, and added to the calendar
- Hardware delivery, after it is ordered
- 2 to 6 weeks
How it is paid
- Deposit 20%
- CAD 43,460 to 66,240
- Discovery 1.3%
- CAD 2,824.90 to 4,305.60
- Data and evaluation plan 8.2%
- CAD 17,818.60 to 27,158.40
- Design approved 1.3%
- CAD 2,824.90 to 4,305.60
- Core features 4%
- CAD 8,692 to 13,248
- Full build 2.6%
- CAD 5,649.80 to 8,611.20
- Training 24.8%
- CAD 53,890.40 to 82,137.60
- Testing and fixes 1.3%
- CAD 2,824.90 to 4,305.60
- Evaluation and red-teaming 16.5%
- CAD 35,854.50 to 54,648.00
- Launch 1.3%
- CAD 2,824.90 to 4,305.60
- Production 8.7%
- CAD 18,905.10 to 28,814.40
- Holdback, 30 days after launch (10%)
- CAD 21,730 to 33,120
- GPU time for training and testing, at cost
- CAD 1,200. Billed up front, at cost, outside the milestones
- GPU server for AI, for the model
- CAD 68,000 to 280,800. Billed up front, at cost, outside the milestones