DatePilot A little more together
This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend * also i have done this with my friend sudharshan also u connect with gemma local models What I Built I built DatePilot, a private AI date optimizer for couples. The idea came from a simple real problem: planning a date for two people is weirdly hard. One person likes quiet cafes, the other wants something outdoors,…
DatePilot is a private AI date optimizer designed specifically for couples, created in response to the challenge of planning dates that satisfy both individuals' preferences. The tool lets two partners input their preferences separately, creating a shared date plan that accounts for factors such as budget, timing, food, travel, and personal dislikes.
To achieve this, DatePilot combines deterministic code with an open-weight model wrapper using Gemma through Ollama. The frontend is built with React, Vite, and Tailwind, ensuring mobile-friendliness for Partner B. The interface is warm and romantic, while the workflow is practical - users create a session, enter preferences, review taste cards, compare overlap, and generate a plan.
The backend is constructed with Python, FastAPI, Pydantic, and SQLite. Rather than letting the LLM decide on prices, schedules, or constraints, the deterministic code checks the real rules, ensuring the total cost fits the budget, venues are open during the planned slot, travel between stops fits the max travel limit, dietary constraints are respected, and excluded dislikes are filtered out. The AI's role is limited to understanding taste, while the deterministic planner handles the math, constraints, and safety checks.
The open-weight model wrapper allows the AI to run locally, enabling immediate deletion of raw uploads and the ability to swap for a hosted open-weight endpoint if needed. Additional APIs, such as OpenStreetMap/Overpass, OSRM, and Open-Meteo, provide venue discovery, route and travel-time estimates, and rain-aware planning, respectively. The venue planner currently supports Indian cities, specifically Tamil Nadu.
This open innovation approach ensures that AI doesn't become a black box that secretly decides the date, instead making AI personal without making it invasive. By using an open-weight model, DatePilot allows taste extraction to run locally, ensuring private taste data remains under the user's control. This combination of open innovation and transparent code makes AI feel different from a normal closed API app.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
