The problem PAZE solves

Paze solves a massive, everyday frustration that millions face in cities worldwide: spotting real problems (potholes, broken sidewalks, damaged bike racks, graffiti, illegal dumping) but having zero realistic way to get them fixed without wasting hours — or giving up entirely.

The Brutal Reality It Fixes

• Time wasted: The average person spends 45–120 minutes just to file one proper city report (finding the right department, filling forms, uploading photos, dealing with errors, following up with no response).
• Low success rate: In many U.S. cities, only ~30–50% of 311-style reports ever receive visible action within 30 days (based on public city dashboards and resident complaints).
• Zero accountability: You submit → radio silence. No timeline, no priority visibility, no way to know if your report even moved the needle.
• Cost to society: Unfixed infrastructure costs cities millions — Denver alone spends ~$15–20 million/year on pothole repairs and sidewalk fixes, yet under-reporting means problems fester longer, accidents rise, and lawsuits pile up.

Paze changes all of that — turning passive frustration into fast, incentivized, trackable action.

What People Can Actually Use Paze For

1. Report any urban issue in <30 seconds
   Open Telegram (already on your phone) → chat with @PazeBot → snap photo/video + quick voice/text note → send. Done.
   → No downloading new apps, no accounts, no long forms.

2. Get real verification & credibility
   AI + ZeroG instantly analyzes and cryptographically proves your media is authentic → no fake reports, no spam, no wasted city staff time.

3. Automatically create & push a formal proposal
   Paze turns your quick report into a structured, on-chain proposal sent to the DAO on ADI Testnet — ready for community review and voting.

4. Influence what actually gets fixed first
   Vote in the DAO (stake tokens or just participate) to prioritize high-impact issues → your voice directly affects which pothole or sidewalk gets repaired next.

5. Bet on (and profit from) when fixes happen
   Every live proposal launches a prediction market: “Will this be fixed by March 15?” or “Under 30 days?”
   → Bet tokens → crowd wisdom sets realistic timelines → accurate predictors earn rewards → cities feel real pressure from public odds.

6. Track everything transparently
   Real-time on-chain status updates via Telegram bot or web dashboard: “Analyzing… Verified… Proposal live… 68% chance fixed in 20 days… Resolved!”

How Paze Makes Existing Tasks Easier, Safer & More Rewarding

Task Today	With Paze	Improvement / Money Impact
Spend 1–2 hours on city website	20–30 seconds in Telegram	Saves ~60–100 minutes per report
Hope your report gets seen	On-chain proposal + DAO voting + prediction market	10× higher visibility & accountability
Zero financial incentive	Earn from correct prediction bets + reputation	Potential $5–$500+ per accurate market call (token rewards)
No follow-up or closure	Automatic notifications + proof-of-fix loop	Closure rate potentially 3–5× higher
Isolated individual complaint	Community DAO + treasury-funded bounties	Turns one voice into collective power; can fund private fixes if city delays

The Hype – Why Paze is a Game-Changer

Imagine if Waze didn't just tell you about traffic — it let you bet on when the jam would clear, rewarded accurate predictors, and used the crowd money to pressure road crews to fix it faster.

That's Paze for civic infrastructure.

• Millions of under-reported issues every year in U.S. cities alone
• $100B+ annual infrastructure repair backlog nationwide
• Prediction markets already move billions in volume on platforms like Polymarket — Paze brings that power to real streets
• Early users could earn real yield by being the first to spot & bet on high-impact fixes
• Cities save money — faster reporting + crowd prioritization = more efficient $15–20M+ annual budgets

Paze isn't just reporting.
It's weaponizing attention, incentives, and blockchain transparency to make cities actually respond.

Report in seconds. Vote with impact. Bet to win. Fix your city.
Paze — People's Waze — is how we finally make "someone should fix this" turn into "we just did." 🚀

Challenges we ran into

Building PAZE was challenging because we were integrating multiple systems at once: a Telegram bot, AI analysis pipeline, decentralized storage (0g), and an on-chain DAO workflow on ADI Testnet. The biggest hurdles were in making these systems talk to each other reliably.

Major challenges we faced

Deploying and stabilizing the Telegram bot

Getting the Telegram bot deployed and running reliably was a major challenge.

We had to handle media uploads, bot event handling, file extraction (frames/images), and make sure the bot could consistently pass the right files into our analysis pipeline.

A lot of issues came from runtime environment differences, file paths, and making sure the bot and the AI/photo analysis service were actually communicating correctly.

0g Storage → ADI DAO transfer workflow

One of the hardest parts was the handoff from 0g/decentralized storage into the ADI DAO proposal system.

It was not enough to just upload data successfully. We also needed to ensure the stored evidence was retrievable, correctly referenced (CID/hash/root), and then mapped into the on-chain proposal metadata without breaking the transaction.

This required extra validation, retries, and normalization before creating proposals on-chain.

End-to-end pipeline synchronization

Our flow spans Telegram intake → AI analysis → storage upload → proposal creation.

Timing/sync issues (especially file watchers, polling, and async processing) caused cases where a file existed but was not processed by the next step.

Debugging this was difficult because failures were often caused by orchestration, not the individual components themselves.

Schema consistency between AI output and smart contracts

AI outputs are flexible, but smart contracts require strict typed values.

We had to normalize fields like severity, issue type, confidence score, and evidence references to avoid failed proposal creation transactions.

Frontend + smart contract integration on ADI Testnet

Wallet connection flow, network checks, chain ID validation, and contract config all needed to be aligned.

Even small config mismatches (RPC URL, contract address, chain ID) could break proposal voting/execution flows.

How we overcame them

Added clear pipeline checkpoints and logging for every stage (bot ingest, analysis, storage upload, proposal creation)

Used processed-file tracking to avoid duplicate runs and identify where the flow was getting stuck

Built a normalization layer to convert AI analysis into contract-safe metadata

Added validation + retry logic for storage retrieval before writing proposal data on-chain

Tested each subsystem independently (Telegram bot, storage, contracts, frontend) before reconnecting them

Biggest takeaway

The hardest part of PAZE was not building any one feature, but building a reliable bridge between platforms, especially Telegram bot deployment and the 0g storage to ADI DAO transfer process. Once those handoffs became stable, the rest of the system became much more dependable.

If you want, I can also give you a shorter hackathon-submission version (more punchy, 5 to 7 lines) for that field.

Use of AI tools and agents

Our AI pipeline is designed to help with issue detection, verification, and proposal preparation before anything goes on-chain.

1. AI-powered media analysis (photo/video evidence)

A user submits evidence (for example through the Telegram bot), such as a photo or video of a damaged road, unsafe rail, or environmental issue.

If the input is a video, the system can extract relevant frames/snapshots for analysis.

The AI agent analyzes the visual evidence and generates structured outputs such as:

issue category/type

severity level

confidence score

impact-related observations

summary description for the DAO proposal

This helps convert unstructured evidence into something the DAO can actually vote on.

2. Authenticity and validation support

AI is also used as a verification layer to assess the submitted evidence before it enters governance.

The goal is to reduce spam/fake submissions and improve trust in reported issues.

The result is a more reliable proposal pipeline where evidence quality is considered before community voting.

3. Proposal generation agent

After analysis, a backend agent converts the AI output into a normalized proposal schema that matches the smart contract requirements.

This includes mapping AI results into fields used by the DAO, such as:

issue type

severity

evidence reference (CID/hash)

confidence/verification score

metadata for proposal creation

This is important because AI outputs are flexible, but smart contracts require strict typed inputs.

How the AI agents work together in the system

PAZE uses a multi-step agent workflow rather than a single “black box” AI call.

End-to-end flow

Telegram bot (data collection agent/interface)

Collects community submissions (photo/video + optional note)

Triggers the processing pipeline

Media processing + AI analysis agent

Extracts relevant frame(s) from video if needed

Analyzes the visual evidence

Produces structured issue metadata and summary

Storage integration agent (0g / IPFS)

Uploads evidence + analysis artifacts to decentralized storage

Returns CID/root/hash references for transparency and auditability

DAO proposal creation agent

Normalizes AI + storage outputs into contract-safe metadata

Submits proposal creation transaction to the ADI DAO smart contract

Frontend + community governance

Community members review proposal details and vote

Prediction market / DAO mechanisms are used to forecast and govern outcomes

Why this AI-agent architecture matters

Instead of using AI just for a summary, PAZE uses AI as an operational bridge between the real world and decentralized governance:

AI interprets evidence

Storage preserves evidence

Blockchain enforces transparency

Community voting decides action

This makes the system faster than manual reporting workflows, while still keeping the process verifiable and community-controlled.

In short

PAZE’s AI tools and agents work together to transform a raw field report into a structured, validated, decentralized proposal that can be reviewed, voted on, and tracked transparently on-chain.