Contact person: Business Client

Phone:Show

Email:Show

Location: Bowie, United States

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"My wearable exports heart-rate, step-count and sleep logs as custom-formatted CSV files. I want a lightweight desktop app that I can run on my Windows PC, drop those files into, and immediately see:

• updated personal baselines for resting HR, daily steps and average sleep
• automatic trend detection against those baselines
• context I can tag on the fly (home, gym, travel, etc.) to enrich future pattern-finding
• clear, friendly recommendations such as “take a short walk” or “drink more water,” rewritten on each new import and, if possible, phrased through OpenAI so they sound naturally personalised rather than robotic

Core expectations
A simple, clean interface is enough: one screen to select/import a CSV, another to review insights and add a quick label. Behind the scenes the app should persist my historical metrics, recalculate baselines with every new file, and compare fresh data to my norms before generating advice.

Because my CSV layout is custom, the first-run flow should let me map my column headers to the fields you expect (heart rate, steps, sleep duration). After that, subsequent uploads use the saved mapping automatically.

Deliverables

• Executable desktop application for Windows 10/11
• Source code with clear comments and a README covering setup, the column-mapping step, and how to supply an OpenAI key (if you choose to include GPT-powered phrasing)
• Simple local datastore (SQLite, JSON or similar) holding historical metrics and tags
• One-click import that updates baselines, detects deviations and outputs refreshed recommendations in under a few seconds

Acceptance criteria
The app parses at least one week of my sample files without errors, shows evolving baselines on each additional import, and produces context-aware suggestions that change when I tag a session “gym” versus “home.”

Python with pandas is preferred." (client-provided description)