Cookwala / How Cookwala helps
Core 0.2 · normative, under review
How Cookwala helps
Three goals, each told with its measured problem, what exists now, what comes next and later, and what we will never claim. Every number carries its label: measured, modelled or assumed.
measured: counted by a named source modelled: from a simulator, with stated assumptions assumed: our estimate, not yet measured
Help end hunger
Hunger has many causes: poverty, conflict, climate shocks, prices and policy. Cookwala's part is real and partial: food that would be thrown away reaching a plate safely, and programs that can count the same way. 733 million people faced hunger in 2023 measured; about 14 % of food is lost before retail and 1.05 billion tonnes are wasted after it measured.
Now
Now: the Humanitarian Profile. A store or farm offers surplus by SMS from a basic phone; a food bank claims it; every handover records a temperature and refuses what is unsafe; a kitchen reports meals; the summary is computed from the documents.
Next
Next: a pre-registered pilot with one program, an independent evaluator and published results, whatever they show.
Later
Later: aggregated, delayed demand and supply signals so less is grown and moved to be thrown away, after a competition-law review.
Never
Never: a claim that Cookwala ends hunger, or a number without its source and label.
Read more: Humanitarian · Farmers · Impact · Pilot protocol
Make people healthier
Unsafe food causes about 600 million illnesses a year measured WHO; adults eat about 11 g of salt a day against a guideline of under 5 g measured WHO 2021. A cooking machine adds new ways to get it wrong, and new ways to get it right every time.
Now
Now: critical control points a device must meet and record (core temperatures, hot holding, cooling, reheating), and rule packs: a reviewable list of food-safety and nutrition rules a program chooses to apply, derived from published guidance that flag sodium, free sugars, fats, fruit and vegetables on a menu, with care rules for children, pregnancy and older adults.
Next
Next: a dietitian and a food-safety officer review each rule pack and sign the review record; the packs say they are unreviewed until then.
Later
Later: national codes replace the defaults where they exist; clean cooking enters the model.
Never
Never: diagnosis, treatment or therapeutic diets. General guidance only; clinician targets stay in the home.
Read more: Health claims policy · Review template · For dietitians and food-safety officers
Put robots to work for people
Robots are learning to move; nobody had written down how to cook. Each maker decides alone what simmer means and when chicken is safe. A shared, checkable definition lets any machine cook any cuisine, and refuse what it cannot verify.
Now
Now: a recipe says what to make, when each step is done and what must never happen; a device dry-runs it before any heat, climbs a sensor ladder: the ways a device may verify a step, best first: a sensor, then a logged estimate, then time, then a person watching for each step, and refuses rather than guesses. Logs leave a home only with consent; recipes keep the cook's name.
Next
Next: a first device running a recipe without editing; conformance reports from makers; a seat for the people served in governance.
Later
Later: certification by independent certifiers; robots that serve people who cannot use a stove today, with the hard case stated plainly.
Never
Never: a machine that guesses a temperature, raises its own safety limits, or learns from a home without consent and credit.
Read more: Try the dry run · For robot makers · Dignity in automated care · Conformance and certification
What we will never claim
No users, partners, pilots or endorsements exist today, and the site says so wherever a reader might think otherwise. The simulators are illustrative. No recipe is field-verified. Safety and privacy are never weaker than Core 0.2. Trust · The rules we write by