MAC.BID Auction Research
An auction research system that compares potential purchases, accounts for costs and tests simulated decisions against closed-lot outcomes.
- My role
- Research workflow, risk-policy design and AI-assisted implementation.
The problem
A low auction price can hide fees, transport costs, uncertain condition and weak resale evidence.
Constraints
- Keep simulated decisions separate from purchases
- Include fees and logistics in the cost
- Treat sparse price evidence conservatively
Current stage
Shadow research
Research output
The system prepares a cost-aware shortlist and records simulated bidding decisions for later comparison with auction outcomes. This makes it possible to inspect the reasoning without treating a suggested purchase as a completed trade.
First-hand accountWorking with uncertain prices
Catalogue information is checked against available price evidence. Fees, tax and transport affect the total cost, while uncertain condition or weak comparisons reduce confidence. The research records a fixed simulated maximum instead of chasing each auction extension.
First-hand accountResearch limits
The workflow presented here does not submit real bids. Its profitability score is an uncalibrated heuristic, not a measured chance of success. Closed-lot history supports further testing; it does not prove resale profit.
First-hand accountView the technical architecture
Conceptual workflow
- 01Public catalogue
- 02Cost and condition review
- 03Simulated maximum
- 04Closed-lot comparison
Decisions
- Enforce budget rules in code
- Record a fixed simulated maximum
- Keep price evidence and uncertainty visible
Evidence
- Catalogue scanning and valuation codeFirst-hand account
- Closed-lot history and counterfactual backtestsFirst-hand account
- Tests for valuation and research workflowsFirst-hand account
Safety boundary
- This case covers research without real bids
- Profitability scores are uncalibrated heuristics, not measured probabilities
- No profit guarantee, account details or private trading history is published
Working vocabulary
Python / Docker / Rule-based scoring / Backtesting / Workflow automation
The basis for this case.
Record reviewed: . This describes the published account, not a new check of the project's runtime.
My role, implementation details and unlinked outcomes are first-hand statements. Public sources support only the claims identified alongside them. A project status or screenshot does not independently verify every result.