great stuff mate! 

looking forward to seeing it in action. two sats tho
1) "While accepting Lightning reduces the fees on my client’s side, it just moves the burden of paying the on-chain fee to my side, effectively making my business a speculator on the on-chain fee market"
This is primarily true for a business/person that only receives payments in bitcoin but primarily spends fiat (tho now with exchanges (and p2p solutions) having LN support even that becomes a lesser problem)

2) "I’ve seen wallets come and go, including the one we built. So I know how hard it is to find a business case. But I feel now is the right time in terms of technology and adoption to make a bet on bitcoin payments"
I've been thinking a lot about this - in properly build trustless projects the business model can't be rent seeking and user capture but an actual service people pay for willingly, not because they don't have a choice. 

And a service people are paying for already is bridges between security models. So a wallet/service that enables me to seamlessly switch between risk models (mints, liquid, lightning and onchain) is something that will make money (boltz.exchange being a clear proof of that). I'm looking very much towards more multi-risk options self-custodial wallets (kudos to  nostr:npub1jugar2agq6369p0l86razavs9shj2p6pscxecevs8j94ap37hkqsjlfc28,  nostr:npub148qm45zettnf6ekgkatnyfadunxwjpu8sy88mjdsgwc5f202d93qmejra7,  nostr:npub1mutnyacc9uc4t5mmxvpprwsauj5p2qxq95v4a9j0jxl8wnkfvuyque23vg that already support at least two options - LN and onchain).