Files
rwo-docs/AGENTS.md
alex wiesner 1f12ba8007 initial commit
2026-03-20 18:48:09 +00:00

91 lines
3.6 KiB
Markdown

# AGENTS.md
## Project overview
This repository is a documentation-first workspace for a startup concept.
The startup thesis is a B2B2C fintech platform that processes or reconciles customer payments, calculates commissions in real time, gives employees fast access to earned commissions, and gives employers better controls and analytics around commission payouts.
## Current strategic thesis
- Start in Brazil.
- Lead with `Pix`-based instant access to earned commissions.
- Use a regulated financial partner for custody, payouts, card issuance, and compliance.
- Treat cards and stored balances as later retention tools, not the initial wedge.
- Treat advances as a later employer-backed feature, not the first product.
- Build employer analytics from payment-linked data.
## What this company is
- commission infrastructure
- a ledger and payout orchestration layer
- an employer workflow product with employee financial utility
## What this company is not
- not a generic neobank
- not a full payroll processor
- not a full CRM
- not a payday lender
## Repository status
- There is no production code yet.
- The repo currently stores strategy and planning documents.
## Key documents
- `docs/core-idea.md` - canonical summary of the product and business model
- `docs/brazil-market.md` - Brazil-specific thesis and go-to-market notes in pt-BR
- `docs/market-memo.md` - competitor map, differentiation, and investor-style memo
- `docs/positioning-and-landing-copy.md` - messaging and homepage copy directions
- `docs/year-one-roadmap.md` - phased roadmap for the first year
- `docs/ideas-for-review.md` - review file for ideas not yet integrated into the core docs
## Language conventions
- Write in English by default.
- Keep `docs/brazil-market.md` in pt-BR.
- Keep fintech terminology consistent: `Pix`, `commission ledger`, `earned commission access`, `employer-backed advance`, `stored balance`.
## Transcripts
- Transcripts of previous sessions are available under the sessions/ subdir, in case extra context is necessary.
## Product assumptions to preserve
- Early architecture should rely on a licensed partner rather than direct custody.
- The safest wedge is real-time access to earned commissions, not general lending.
- The payment event is the source of truth for commission creation whenever possible.
- The employer buyer needs controls, auditability, and analytics, not just a payout rail.
- The employee experience matters because faster access and later card usage can improve retention and platform utility.
## Reference competitors and adjacent players
Use this current market framing unless new evidence clearly changes it:
- `DailyPay`
- `Branch`
- `PayQuicker`
- `Hyperwallet`
- `CaptivateIQ`
- `Caju`
- `Dock`
These companies validate parts of the stack, but the current thesis is that few combine payment-linked commissions, worker liquidity, employer controls, and Brazil-native rails into one product.
## Prioritization guidance for future agents
- Prioritize docs and plans that sharpen the Brazil-first wedge.
- Keep the distinction between `earned access` and `credit` very clear.
- Prefer practical operator language over generic fintech buzzwords.
- If you extend product plans, sequence features conservatively: ledger and instant payouts first, analytics and stored balance second, cards third, employer-backed advances later.
## Open areas worth exploring later
- deeper regulatory mapping for Brazil launch structure
- partner comparison across BaaS, payment processors, and card issuers
- pricing design for employers versus workers
- reserve rules for refunds, disputes, and chargebacks
- target vertical selection and design-partner outreach strategy