MarketLens

Waze for street market prices — price transparency for the informal economy.

Comment

Inspiration

Informal and street markets across developing countries operate with limited price transparency, making it difficult for buyers to know what a fair price should be. With over 2 billion workers and nearly $10 trillion in annual transactions flowing through informal economies, there is a massive lack of pricing infrastructure for everyday consumers.

MarketLens was built to close this information gap by providing real-time, data-driven pricing intelligence. The platform empowers buyers with contextual market insights so they can negotiate and purchase with confidence.

What it does

MarketLens is a pricing intelligence platform for informal markets that helps users quickly determine whether a quoted price is fair.

The platform:

  • Aggregates crowdsourced price observations from local users
  • Combines global macroeconomic indicators with regional commodity baselines
  • Calculates an Informal Inflation Index (III) and related market signals
  • Displays insights using maps, heatmaps, and market dashboards
  • Provides AI-generated guidance including price context, fairness scoring, and negotiation suggestions

Users can search items, scan products using images, submit prices, and instantly receive analysis comparing local market pricing to global benchmarks.

How we built it

We built MarketLens as a full-stack platform using a mobile-first architecture.

Backend

  • Built with Python and Flask to handle data aggregation and intelligence computations
  • Integrated World Bank economic indicators, FAO commodity price datasets, and humanitarian disaster feeds
  • Used statistical modeling (z-score anomaly detection, volatility scoring, trend acceleration metrics) to generate pricing signals
  • Integrated Claude AI to generate contextual market analysis, item identification, and negotiation scripts

Frontend

  • Developed using React Native and Expo for cross-platform mobile and web support
  • Implemented React Native Web to deliver a consistent experience across devices
  • Built interactive market dashboards, search tools, and submission flows
  • Used maps and card-based visualizations to make pricing patterns easy to understand

Data & Storage

  • PostgreSQL (Neon) stores all price observations and intelligence metrics
  • Implemented caching and dataset preloading to reduce API latency

External Intelligence

  • World Bank APIs provide CPI and food inflation data
  • ReliefWeb disaster feeds dynamically adjust supply shock multipliers
  • FAO producer price datasets establish historical price floors

Challenges we ran into

The biggest challenge was balancing product scope with usability while still demonstrating meaningful real-world impact.

Technically, integrating multiple live data sources introduced latency and synchronization challenges. We solved this by:

  • Preloading critical datasets on app launch
  • Using caching layers on the backend
  • Parallelizing baseline calculations using multi-threaded processing

Building the AI and economic analytics pipeline was also complex because we needed to present quantitative insights in a simple, user-friendly format without overwhelming users with raw data.

Another challenge was aligning product design with functionality — ensuring maps, price intelligence metrics, and AI guidance worked together seamlessly rather than feeling like separate features.

Accomplishments that we’re proud of

We are proud to have built a functional pricing intelligence layer for informal economies — an area that has historically lacked structured data infrastructure.

Key accomplishments include:

  • Successfully merging crowdsourced pricing with global macroeconomic benchmarks
  • Designing and deploying real-time market intelligence metrics like III, volatility, and RID
  • Building a fast, mobile-first analytics experience despite working with large datasets
  • Integrating AI-driven contextual guidance directly into consumer decision workflows

What we learned

We learned how to design systems that combine economics, data science, and product design into a single coherent platform.

Key technical learnings included:

  • Integrating multiple external APIs into a unified analytics pipeline
  • Building statistical anomaly detection models for price validation
  • Optimizing mobile data delivery through caching and precomputation
  • Designing AI prompts to produce actionable, culturally contextual guidance

We also learned the importance of shipping a focused, polished product rather than trying to include too many features.

What’s next for MarketLens

Future plans include:

  • Expanding geographic coverage to more global informal markets
  • Improving AI product recognition accuracy through larger training datasets
  • Adding predictive pricing models for future market trend forecasting
  • Growing the crowdsourced contributor network to improve data reliability
  • Building B2B intelligence tools for NGOs, humanitarian agencies, and supply chain analysts

Our long-term vision is to create global pricing infrastructure for informal economies, helping improve transparency and economic efficiency for billions of consumers.

Built With

Share this project:

Updates