Confío: Web2-to-Algorand Consumer Onboarding Infrastructure (LATAM)

Governance Discussions xGov Proposals
1

Confío: Web2-to-Algorand Consumer Onboarding Infrastructure (LATAM)

Amount Requested: 250,000 ALGO
Category: Mass Adoption / Real-World Utility
Type: Retroactive + Operational Continuation
Stage: Live on Algorand Mainnet
Repository: GitHub - caesar4321/Confio: Latin America's PayPal (MIT License, 1,039 commits)

[3-min video pitch]: https://youtube.com/shorts/cdtfDAIn_C4?feature=share

1. What We Have Already Delivered to Algorand

Confío is not a proposal-stage idea. It is a deployed, open-source consumer wallet operating on Algorand mainnet today.

Delivered infrastructure (live):

  • React Native mobile app (iOS + Android, App Store + Google Play)

  • Django + GraphQL backend (AWS)

  • PyTeal/TEAL smart contracts deployed on Algorand mainnet

  • Sponsored atomic group transactions — zero gas for users, ever

  • Client-side private key generation encrypted in Google Drive / iCloud — no seed phrases

  • Phone number → Algorand address mapping for WhatsApp-like UX

  • cUSD: 1:1 USDC-a backed stablecoin, live on Algorand mainnet

  • 3-of-5 multisig governance on critical contracts

All contracts and integration logic are open source under MIT license.

Verified traction (on-chain):

  • 5,500+ fully registered users (Google/Apple login + SMS verified)

  • 70+ funded users with real USDC-a deposited on-chain

  • 2,000+ cUSD minted and circulating on Algorand mainnet

  • 430,000+ Spanish-speaking TikTok followers (founder distribution channel)

  • 12,000+ Telegram community members

  • Algorand Foundation Accelerator participant

This is direct Web2 onboarding into Algorand wallets — not exchange users, not airdrop hunters. All activity is publicly verifiable on-chain.

2. Why This Infrastructure Matters

The Problem Is Not Technology

Algorand has world-class technical properties: ~3.5 second finality, sub-cent fees, zero chain halts, native USDC-a. What it currently lacks is a consumer distribution layer in the regions where dollar demand is most acute.

Latin America has 650 million people. Argentina, Venezuela, Bolivia, and Colombia collectively represent one of the largest real-world stablecoin markets on earth — not because of crypto adoption, but because of structural economic failure.

In 2001, Argentina froze every bank account in the country overnight. Families woke up and could not access their savings. This event — the “Corralito” — did not just destroy wealth. It destroyed a generation’s trust in custodial financial systems.

That trauma is not historical. It actively shapes financial behavior today. It is why Argentines refuse to keep meaningful savings in MercadoPago, Lemon, or any app that holds their money for them. Self-custody is not blockchain ideology for these users. It is basic risk management learned from lived experience.

The Structural Gap No One Has Filled

Every crypto project targeting LATAM makes the same mistake: they build for crypto users.

They offer trading interfaces, yield products, seed phrase onboarding. They optimize for people who already have crypto and know how to use it.

The people who most need access to stable dollars are not crypto users. They are WhatsApp users. They send money via phone number, not wallet address. They will not install MetaMask. They will not write down 24 words.

Confío is built for them.

Why Confío Is the Dark Horse

In LATAM, trust is not a feature. It is the distribution channel.

The reason no Web3 project has cracked mass adoption in LATAM is not technology. It is confianza — the human kind. People will not deposit life savings into a protocol without a face, a track record, and a real-world presence.

Confío is built around this insight:

  • Non-custodial architecture → blockchain trust (your keys, your funds, always)

  • Real-name founder with 430K+ LATAM followers → human trust (a face, accountability, skin in the game)

  • Spanish-first UX, zero crypto complexity → cultural trust (built for LATAM, not translated for it)

We have 5,500 registered users and 12,000 Telegram members not because of a token incentive. We have them because people in Latin America trust the founder, and the founder built them a non-custodial wallet on Algorand.

That is a distribution moat that cannot be bought.

3. Technical Architecture

User (Web2 UX — no crypto knowledge required)

         ↓

Google / Apple Login

         ↓

Client-side key generation

→ Encrypted in Google Drive (Android) / iCloud Keychain (iOS)

         ↓

Algorand address created automatically

         ↓

Phone number → address mapping (backend)

         ↓

Sponsored atomic group transaction

→ Confío pays gas. User pays nothing.

         ↓

Algorand mainnet settlement (~3.5 seconds)

         ↓

User sees: "Sent $10 to María ✓"

Users never encounter seed phrases, gas fees, blockchain explorers, or trading interfaces. Yet every transaction settles directly on Algorand.

Open source components (MIT licensed):

  • /contracts — PyTeal/TEAL smart contracts (multisig, cUSD minting/burning)

  • /blockchain — Algorand transaction layer

  • /conversion — ALGO↔USDC-a swap logic via Tinyman

  • /apps — React Native mobile app (iOS + Android)

  • /payments, /send, /payroll — core financial flows

1,039 commits. Fully open source since day one.

4. Retroactive Contribution to the Ecosystem

Confío has already delivered to Algorand:

  • 1,039 open-source commits

  • Production PyTeal smart contracts on mainnet

  • Sponsored gas model implementation at consumer scale

  • Web2 authentication → non-custodial wallet automation pipeline

  • Real stablecoin transaction volume on Algorand mainnet

  • 5,500 new Algorand wallets created via Web2 onboarding

The infrastructure layer required to onboard non-crypto-native LATAM users exists. It runs on mainnet. It creates Algorand wallets. It processes real transactions.

5. What This Grant Enables

The requested 250,000 ALGO is not for speculative development.

The infrastructure is already built and live.

This grant allows the existing system to operate and scale through the critical Argentina launch window — converting deployed on-chain infrastructure into measurable consumer adoption over six months.

This is operational continuation, not concept funding. We are a two-person team at minimal burn. The grant covers the runway needed to complete what we have already built.

6. Measurable On-Chain Commitments

All metrics are publicly verifiable on Algorand mainnet.

By Month 2:

  • ARS on/off-ramp live

  • 150+ funded users with on-chain USDC-a balances

  • Public real-time dashboard with on-chain reporting

By Month 4:

  • 500+ funded users

  • $100,000+ cumulative USDC-a processed on Algorand

  • 10,000+ additional sponsored atomic transactions

By Month 6:

  • 1,000+ funded Algorand wallets

  • Infrastructure ready for Colombia/Mexico Phase 2

  • Full milestone report published to xGov forum

We will report progress publicly at each milestone. Every number is auditable on-chain. We are not asking for trust — we are offering proof.

7. Governance Responsibility

We are requesting 250,000 ALGO with 5,500 registered users, 70+ funded users, live mainnet contracts, and 430,000+ real LATAM followers who have never been offered a non-custodial wallet they could actually use.

We deliberately request less than the maximum allocation. Our approach is to prove measurable consumer adoption first, report transparently, and request additional ecosystem support only after delivery.

About the team

Julian Moon — Founder & CEO

• Boots on the Ground: A Korean full-stack developer who left his home country to live nomadically across Latin America for the past 5 years. He builds “not from a distant office, but living here, sharing the exact same economic challenges as the users.”

• Zero-CAC Distribution: Built a Spanish-speaking TikTok audience of 430,000+ followers. Instead of burning millions on ads, he uses a viral, “Zero-CAC” (Customer Acquisition Cost) strategy to convert cultural attention directly into Algorand adoption.

• Execution & Bridge: Host of Crypto News Korea (120+ episodes) and successful founder of multiple cross-cultural digital platforms. He connects high-level tech execution with Latin American realities.

Susy Ramirez — Content Lead & User Acquisition

• Drives the distribution strategy, video production, and community engagement.

• Crucial to our user acquisition funnel, directly managing the 12,000+ Telegram community and converting Web2 traffic into funded on-chain users.

Open Source & Lean Execution We operate as a highly efficient, lean team. Our entire infrastructure and contracts are fully open-source (MIT Licensed) from day one.

Additional Info
Proof of Work & Community Links:

3-Min Video Pitch (Recorded in Buenos Aires): https://youtube.com/shorts/cdtfDAIn_C4?feature=share

App Website: https://confio.lat

GitHub Repository (1,039 commits): GitHub - caesar4321/Confio: Latin America's PayPal

Founder’s TikTok (430K+ Followers): TikTok - Make Your Day

Telegram Community (12K+ Members): Telegram: View @confio4world

Final Note to xGovs: We welcome technical questions, on-chain verification requests, and governance scrutiny. Confío is not asking for funds to build an idea. We are asking for the exact operational runway needed to scale an infrastructure that is already live and running on Algorand mainnet today.

2

I think, proposals should stand on their own and not try to justify their ask relative to other proposals.

Regarding the proposal itself: this is a fairly new project with impressive social profiles and some early traction, as per your wording:

  • 5,500 new Algorand wallets created via Web2 onboarding
  • 70+ funded users with real USDC-a deposited on-chain,
    I would presonally want to see the stack be used for a bit longer before i could make an informed decision about it’s value for the ecosystem.

Proposal is compliant with TnC so i personally have no problem voting on it as such when the time comes.

3

Hi Simon, thank you for the thoughtful feedback.

Noted on the Lute comparison — I’ll remove that framing. You’re right that the proposal should stand on its own merits.

On the “fairly new” concern: totally valid. The infrastructure has been live since late 2025, but the Argentina/LATAM complete on/off-ramp (the key activation layer) is indeed in its final integration phase now. The 5,500 registered users and 70+ funded users represent what we’ve achieved without a working fiat rail — once that’s live, the conversion funnel opens properly.

Happy to answer any specific technical or governance questions during discussion period.

Julian

4

Thanks for the swift response! Your explanation makes sense and as i said i presonally see the traction there. Any sort of innovation on algorand is welcome in my mind and xgov is here to support it.

5

Happy to support projects that are live and having good potential to grow.

6

I think the project has a good approach to tackling a valid problem!
I’d be curious to learn more about what is the role and benefit of cUSD compared to USDC, which is backing it? Why not use directly USDC?

From the repo, I see many references to the Sui blockchain. What is the project’s relationship with Sui? Can you share more on the project’s history?

I also see the contracts are written in PyTeal (which is outdated tech). Why was this decision made? If I understand correctly, the project launched on Algorand in 2025?

There is also mention of a CONFIO token in the repo. What’s its role and is that also part of the grant request?

Also, would you mind sharing any references from your participation in the Algorand Foundation’s Accelerator program? What was the outcome and/or what feedback did you receive there?

Lastly, due to the video and the proposal description, I would like to point out that the xGov program is currently for retroactive funding only. While the work itself may qualify for funding, I would deem the references to how the funds will be spend and the promises for future deliveries irrelevant for the proposal because of this.

7

Hi, thank you for the deep dive into the repo and the thoughtful questions! I love these technical and historical inquiries. Let me walk you through exactly why and how we built this.

1. Why cUSD(Confío Dollar) instead of direct USDC?

Three reasons.

First, branding — our target users don’t distinguish between USDT and USDC. If we’re doing financial education anyway, we’d rather build equity in our own brand than someone else’s.

Second, ecosystem optionality — having our own stablecoin layer gives us future flexibility for LATAM-specific financial products that USDC alone wouldn’t support.

Third, yield distribution — longer-term, as cUSD circulation grows, we’re exploring a USDC-backed cUSD model where reserve yield gets partially distributed back to users holding cUSD in-app (similar to how some neobanks offer savings rates). This isn’t live yet and isn’t part of this grant, but it’s why the wrapper architecture exists. We’re looking at specifically mTBILL. This allows us to generate yield (e.g., 4%), retain 1% for Confío’s sustainable revenue, and distribute 3% directly to our users holding cUSD. cUSD is the foundation for this LATAM ecosystem.

2. The Sui/Aptos Traces & Why We Chose Algorand

This is actually my favorite part of our startup journey. When choosing a chain, I had 3 absolute requirements for Web2 onboarding:

  1. Social Login (no seed phrases),

  2. Chain-level sponsored gas,

  3. Fast finality.

Initially, Sui’s “zkLogin” marketing seemed perfect. But the reality was different: the open docs were insufficient to build the prover without enterprise-level direct support from the Sui Foundation, which we couldn’t get as an early startup. We pivoted to Aptos, which had a similar “Keyless” feature. We still faced integration blockers, so I submitted the Confío project to an Aptos Hackathon just to get a direct contact. That Hackathon submission ended up on the radar of Ram from the Algorand Accelerator. Ram reached out and pitched Algorand. I realized Algorand was actually the superior choice: we could sponsor gas beautifully via Atomic Group Transactions, build our own client-side cloud key management, and most importantly for a financial app, Algorand has zero downtime and instant deterministic finality (no forks). We made the hard pivot to Algorand and never looked back.

3. Why PyTeal? (Launched Late 2025) You’re correct that PyTeal is older relative to newer tooling. The decision was purely pragmatic and based on developer familiarity at our initial build stage, speed of prototyping during the MVP phase, and the proven stability/auditability of AVM-based PyTeal contracts. Our priority was getting a safe product into users’ hands in late 2025. Future iterations may migrate to more modern tooling like Puya as the ecosystem evolves.

4. The CONFIO Token To be absolutely clear: The CONFIO token is not part of this grant request. It is referenced in the repo as a utility/reward token for broader ecosystem planning (like referral rewards). The 250,000 ALGO proposal strictly concerns the live wallet infrastructure, on-chain cUSD deployment, and scaling our user onboarding. No grant funding is related to speculative token mechanics.

5. Algorand Accelerator & Retroactive We participated in the Algorand Startup Accelerator and worked closely with Ram and Will. The outcomes were purely mentorship-oriented: technical feedback on our development architecture, ecosystem alignment, and GTM refinement for LATAM. We received no direct funding from the Foundation. And you are 100% right on the retroactive point. While we included our expansion plans to show why our infrastructure is valuable, we fully expect and respect being evaluated solely on the sunk costs, the deployed mainnet contracts, the 5,500+ Web2 wallets we have already delivered.

Thank you again for the thorough review. I’m happy to provide any further technical details!

8

Confío xGov proposal — close to quorum, need final votes

Our proposal already has strong majority support, but we are slightly short on quorum.

If you’re an xGov participant, your vote can directly decide the outcome.

Vote here (1 min):

9

As the proposal didn’t reach the quorum the first time and it is now ~1.5 month since the initial proposal application, can you share how have your metrics changed in the meantime (in particular any on-chain metrics, which can be independently verified), or how has the project progressed?

10

Thanks for the question — happy to share a concrete update.

As this is a retroactive proposal, I did not modify the original post. Instead, here is a transparent update on progress over the past ~1.5 months.

1) User growth and demand validation

Confío has grown from 5,500+ to 6,661 phone-complete users (Google/Apple login + SMS verified).

Current distribution:

  • Venezuela: 1,428 users (largest total user base)
  • Argentina: 939 users (now dominant in activity)
  • Bolivia: 287 users (early but growing)

This shows that demand is already validated, especially in Venezuela where stablecoin usage is driven by real economic need.

2) Founder-led market shift (Venezuela → Argentina)

Over the past months, the founder has been physically operating in Argentina, aligning distribution and user acquisition on the ground.

Following this, we observed a clear behavioral shift:

  • Argentina overtook Venezuela in MAU (485 vs 145)
  • Activity concentrated in markets where compliant fiat rails are feasible

This is not organic drift — it reflects deliberate alignment between:
distribution, geography, and infrastructure readiness.

3) Why on-chain growth is currently bottlenecked

The main constraint is not product or demand — it is fiat infrastructure.

For crypto-native users, card on-ramps may be sufficient.
For mainstream LATAM users, they are not.

Users rely on local rails:

  • Argentina: bank transfer / CVU / Alias / Mercado Pago
  • Peru: Yape
  • Colombia: Nequi
  • Bolivia: QR-based rails

Equally important: off-ramp

If users cannot reliably exit back to local currency, the wallet becomes a one-way system rather than a usable financial tool.

This is why partners like Koywe are critical — they bridge local fiat systems to Algorand-native USDC.

4) Infrastructure progress since submission

Since the proposal:

This indicates real onboarding intent before incentives.

5) On-chain status and verifiability

To clarify what can already be independently verified on-chain:

  • Confío has 6,661 phone-complete users, and each user corresponds to a unique, non-custodial Algorand address generated client-side (1:1 mapping)
  • We have recorded 95 USDC deposit events on Algorand mainnet (publicly traceable via Algorand explorers)
  • cUSD mint/burn activity is live and tied directly to these deposits via our smart contract layer

All of this activity is publicly traceable on Algorand, as the system operates fully on mainnet with:

  • user-owned wallets (no custody)
  • sponsored transactions
  • on-chain stablecoin conversion flows

Important note for transparency:

  • cUSD circulation is temporarily lower than at the time of the original post
  • we used a portion of capital provided by early supporters to post the 7,500 ALGO bond required for this xGov proposal.

This was a deliberate operational decision, not a drop in system usage.

6) Near-term update

The primary constraint on further on-chain growth is the activation of compliant fiat rails.

Once Koywe production integration is completed, we expect a direct increase in:

  • funded wallets
  • USDC deposits
  • cUSD mint activity

We are currently in the final stage of production readiness with the Koywe integration.

We will share a follow-up update with expanded, independently verifiable on-chain metrics once this layer is live.

11

Thanks for this great update! Would you mind sharing additionally some of the relevant transactions that can be independently verified on-chain?

12

Thanks — sharing a few representative Algorand mainnet examples below.

cUSD references:

xGov bond:

Representative USDC deposit examples:

Representative cUSD mint examples:

Representative user transfer examples:

Representative cUSD burn examples:

These are only a sample, but they should make the current on-chain activity independently verifiable.

13

A big yes from me:

  • The codebase is solid
  • The mission is well defined
  • The impact is obvious for an startup
  • Repo, website and socials seem to be well aligend with mission
  • Roadmap is sound

The only two questions (not related to value of the proposal) are
1- why cUSD and not using USDC directly?
2- Please elaborate a bit more on architecture of web2 OAuth to Web 3 onboarding (i see it is using Firebase Authentication and JWT verification but a little more explaination will be awesome)

14

Thank you for the thoughtful questions.

  1. Why cUSD (Confío Dollar) instead of using USDC directly?

Three reasons.

First, branding and user trust. Our target users generally do not distinguish between USDT and USDC at the brand level. Since we are already doing the work of user education, we would rather build that trust and familiarity into Confío’s own financial layer instead of outsourcing the user relationship to a third-party brand.

Second, ecosystem optionality. cUSD gives us a programmable stable-value layer tailored to LATAM use cases. Using raw USDC alone would limit our ability to build localized financial products, incentives, and user-facing experiences that sit on top of the dollar balance.

Third, yield distribution design. Longer term, we are exploring a USDC-backed cUSD model where reserve yield can be partially returned to users holding cUSD in-app, similar to how some neobanks offer savings yield. This is not live yet and it is not part of this grant, but it is a key reason the wrapper architecture exists. The idea we are exploring is to back cUSD with USDC reserves allocated into instruments such as mTBILL, generate yield on those reserves, keep a portion as sustainable protocol revenue, and pass the rest back to users holding cUSD. For example, if reserve yield were 4%, Confío could retain 1% as revenue and distribute 3% to users. In that sense, cUSD is not just a wrapper for branding purposes, but the foundation for a broader LATAM-native stablecoin ecosystem.

  1. Can you elaborate more on the architecture of the Web2 OAuth → Web3 onboarding flow?

Yes. The key principle is that we use Web2 OAuth for identity and onboarding convenience, but not for custody.

A user signs in with Google or Apple, and Firebase Authentication standardizes that flow. The mobile client gets a Firebase ID token, and our backend verifies that token server-side to extract trusted identity data such as the Firebase UID, provider, and basic user claims. Once verified, the backend creates or resolves the Confío user and issues Confío JWTs for application authorization.

From there, the system is intentionally split into separate layers:

First, identity and authorization.
Firebase proves who the user is. After verification, our backend issues Confío JWTs that embed explicit account context, such as whether the user is operating in a personal or business account, the account index, and the relevant business scope when applicable. This is important because it means the client does not simply pass an account ID and ask the backend to trust it. Instead, the backend enforces account scope from the JWT itself. For business accounts, we also validate permissions against the employee-role relationship in our database before sensitive actions are allowed.

Second, wallet custody.
The backend never stores or controls the user’s signing key. Wallet material is created and managed on the client side only.

In the current architecture, after login the app checks whether a wallet master secret already exists for that user. If it does, the app restores it locally and derives the appropriate wallet/account keys from it. If it does not, the app generates a new high-entropy master secret on-device. That secret is then stored in secure OS storage and becomes the root from which the user’s wallet/account keys are derived.

Third, backup and recovery.
This is what makes the system usable for mainstream users without sacrificing self-custody.

  • On iOS, the app uses secure OS storage and iCloud Keychain as the native safety layer.

  • On Android, the app uses secure local encrypted storage.

  • For cloud recovery and cross-device portability, we support encrypted Google Drive backup using the appData scope, so backups are isolated from the user’s normal Drive files.

  • The Google OAuth access token used for Drive backup is kept on the client and is never sent to our backend.

That separation matters a lot: Firebase ID tokens are used for backend authentication, while Google Drive OAuth access is used only client-side for encrypted backup/recovery. Identity access and cloud backup access are intentionally not treated as the same thing.

Operationally, the flow looks like this:

  1. The user signs in with Google or Apple.

  2. Firebase returns a verified identity token.

  3. Our backend verifies that token and issues Confío JWTs for authorized app access.

  4. The client checks secure local storage and available backup sources for the user’s wallet master secret.

  5. If a secret already exists, it restores it locally. If not, it creates a new one locally.

  6. The client derives the wallet/account keys from that secret.

  7. The backend is updated with the public wallet address only.

  8. All signing remains on the client device.

We also designed the backup system to support roaming and recovery more safely. Instead of relying on a single mutable backup file, the app maintains a manifest/index of wallet backups and stores encrypted wallet backups as distinct files. That lets us support multiple devices and restoration choices without simple overwrite collisions. In cross-platform cases, users can explicitly restore the backup they want rather than silently replacing wallet state.

For additional security, sensitive actions such as enabling cloud backup or restoring a backup can be gated by device biometrics. On the authorization side, JWT-based account context helps prevent account confusion or IDOR-style issues when users switch between personal and business accounts.

So in short:

  • OAuth/Firebase handles identity.

  • Confío JWTs handle authorization and account context.

  • The user’s device handles custody and signing.

  • Google Drive / iCloud Keychain improve recovery and roaming without giving the server control over funds.

This architecture allows us to onboard non-crypto-native users while preserving true self-custody - which is critical in LATAM markets where trust in custodial systems is structurally low.

15

Awesome! Best of luck for this proposal! I fully support this and find it impactful for Algorand!
100% deserves the asked amount and even more IMHO. Thank you for your contributions @julianmoonluna

16

Thank you for your support, @emg110 !!