120 pounds sent.
112 accepted. The difference stays visible.
Help a pantry fill its next food gap with what a neighbor can spare. Follow one real application workflow from an explained match to an honest receiving record.
2 minutes 51 seconds. English captions are visible in the video. All pantry names and records are fictional. The narration is synthetic, not a recording of Shivam's voice.
Built to follow what actually happened.
Pantry Relay protects local reserves, commits stock when a transfer is reserved, records each handoff, and counts the food the receiver accepts. A partial receipt leaves the remaining need visible. Accounts, network permissions, persistent records and exports support the complete workflow.
The next step is a supervised pilot with one coordinator and three nearby pantries. The proposed $149 monthly network price is a hypothesis. No customer, nonprofit partnership or field impact is claimed.
Read the full narration
Picture a food pantry opening tomorrow. There is rice on the shelf, but hardly any produce. A nearby pantry has produce it could share. Someone still has to connect those two facts, arrange the handoff, and find out what actually arrived.
This is Pantry Relay, created by Shivam Gupta with AI assistance. It helps an approved pantry network fill upcoming food gaps with what its neighbors can spare. These are fictional pantries, and every number in this demonstration is sample data.
Cedar Grove has produce above the stock it needs for its own visitors. Eastside needs produce for its next service. In the relay planner, I can see a proposed transfer of one hundred and twenty pounds, with an explanation of why it fits.
The planner checks reserves, restrictions, storage, receiving space, use-by cutoffs and vehicle capacity. Distances are straight-line estimates, so a person still confirms the actual trip. When I reserve the transfer, that stock is committed. Another coordinator cannot promise those same pounds again.
Next comes the receiving pantry's acceptance, pickup and arrival. The app records the handoffs and requires temperature checks when the food needs them. A driver has a simpler delivery view; coordinators handle acceptance and the final receiving record.
Here is the part that matters. We sent one hundred and twenty pounds, but eight pounds arrived damaged. I enter one hundred and twelve pounds accepted and explain the difference. The report counts one hundred and twelve. The missing eight remain visible against the upcoming need.
The receiving pantry's food is protected from being offered straight back out. A failed delivery counts as no food received. A receipt confirmed after the service deadline gets no credit for that missed service. The records follow what happened, including the inconvenient parts.
This is a working application with accounts, separate networks, permissions and persistent records. We test stock protection, permissions and the complete handoff in automated backend and browser checks. It runs without paid AI or mapping APIs. The next question is whether it makes a coordinator's day easier.
Our proposed price is one hundred and forty-nine dollars per network per month for up to ten sites. That price is a hypothesis. We want to test it with one coordinator and three pantries, measuring food gaps, staff time, transport effort and accepted pounds.
Pantry Relay makes food already in the community easier to share, with a record of what reached the next pantry. Our goal is a useful service that operators choose to keep.