Budget: 27000 UAH Deadline: 12 days
We have a practically ready solution for this task, we can quickly adapt and launch it - I suggest discussing it here, I am available ))
Regarding the cost - the estimate is 28,000-65,000 UAH for the first stage of corrections, timeframe 5-12 working days. I will provide an exact estimate after reviewing the code, formulas, and 3-5 wallets where there is the greatest discrepancy with Nansen. If it is necessary not just to tweak the formulas, but to build a complete transaction classification model, the budget may be higher - it's better not to play guessing games.
We have experience with similar tasks involving crypto statistics, wallets, transactions, prices at the time of the deal, financial metrics, and reconciliation with benchmark sources. We work with Moralis at the level of native ETH transfers, ERC-20 transfers, normalization of swaps, fees, transfers, and atypical wallet actions. With Nansen, the key nuance is to restore the rules for classifying operations, as the difference often lies not in the mathematics, but in what is included in realized PnL. With DeFiLlama, we can work either as a source or a control layer for prices, but we need to agree on which source will be primary.
I see the implementation as follows:
- We take several problematic wallets and periods
- We break down operations into buy, sell, transfer, fee, mint, burn, bridge, airdrop, LP
- We compare each transaction with Nansen and find the reason for the discrepancy
- We adjust the PnL, ROI, Win Rate formulas
- We add a test table to ensure new changes do not break already verified cases
From you, we need access to the repository, examples of wallets, comparison periods, and screenshots or exports from Nansen.
Questions:
- Which networks are currently supported - only Ethereum or also Base, Arbitrum, BSC, Solana?
- Is a near-complete match with Nansen required, or is a slight deviation allowed due to a different price source?
Similar examples from Ingello:
- https://business.ingello.com/vorfahr - automation with complex data, rules, and system logic
- https://business.ingello.com/fractal - architecture of automation, agents, and multi-step processes
- https://systems-fl.ingello.com/ua - main page of the team for Freelancehunt