DEV Community

Cover image for helloTali Overview Plan
Harry Lo
Harry Lo

Posted on

helloTali Overview Plan

#healthtech #ai #mentalhealth #softwareengineering

A compassionate care, non-medical, privacy-first app concept (Mood Check-ins • Education • Screeners • Referral)

Background

After publishing my posts about helloTali, a number of people reached out privately with thoughtful suggestions—thank you. A recurring theme was a simple three-pillar structure: Education, Screeners & Monitoring, and Referral & Next Steps. I’ve organised and summarised the key ideas here so I can move forward in a structured way.

Previous Posts

Core concept

helloTali is my non-medical, privacy-first hobby project. The first version will focus on a simple mood check-in to help people reflect on how they’re feeling over time.

After that, the second version of helloTali will follow a suggested direction: a non-medical “compassionate care” app built around a simple three-pillar structure:

1) Education — help the general public learn mental health concepts in plain language

2) Screeners & Monitoring — support lightweight self-reflection over time (mood / stress patterns)

3) Referral & Next Steps — guide users to appropriate support (what kind of professional to see, where to go)

helloTali is not meant to diagnose, treat, or replace clinicians.

The Three Pillars

Pillar 1 — Education

Goal: Make mental health concepts clear, simple, and supportive, without overwhelming people.

What this includes

1) Plain-language explainers

  • “What does a diagnosis mean (and what it doesn’t mean)?”
  • “Fight / flight / freeze / Fawn (and why your body does this)”
  • “Common conditions explained in a neutral, non-clinical way”
  • “What different professionals do (psychologist vs counsellor vs GP etc.)”

1) Micro-lessons / templates

  • Short lessons that are easy to read
  • Consistent structure and tone (so it doesn’t feel like random blog posts)

Example content style (from feedback)

A draft “What is a diagnosis?” explanation from a psychologist is a good example of the tone/style the education pillar needs: plain language, reassuring, and non-stigmatising.

Key design principle

The content already exists in trusted places—the critical part is presentation:

  • short, structured
  • user-friendly
  • avoids jargon
  • avoids “diagnosing” language

Pillar 2 — Screeners and Monitoring

Goal: Offer structured self-checks and trend tracking.

Two levels (recommended)

1) Light daily check-ins (optional)

  • “How are you feeling today?” mood slider + tags
  • Optional short text note (reflection)

1) Weekly screener (validated)

  • Option A: PHQ-9 + GAD-7 (short, widely used)
  • Option B: DASS-21 (depression/anxiety/stress; needs careful licensing check)

Important constraint

For anything “validated”, the safest approach is:

  • ask standard questions
  • use fixed response options
  • apply standard scoring

(Instead of free-form AI interpreting text into scores.)

Open question to resolve early

Confirm software usage/licensing for DASS-21 and any commercial constraints. If unclear, start with PHQ-9 + GAD-7.

Pillar 3 — Referral and Next Steps

Goal: Help users find appropriate support when they want it.

What “Referral” means here (practical)

This is not “medical referral.” It’s signposting + directory:

  • “Who should I talk to next?”
  • “What can each professional do / not do?”
  • “Pros / cons of each option”
  • “Find a clinician / counsellor directory”

Suggested starting point

Start the directory with counsellors first (often less bureaucracy), and expand later.

Phased approach

1) Phase 1 (simple): “Next steps” page + links + guidance

2) Phase 2: curated directory (invite a small group to join)

3) Phase 3: self-serve signup + verification process + location filters

Suggested MVP Roadmap

Keep the early build small and safe.

MVP 1 (First version) — Mood Check-in + Trend Reflection

  • Daily mood check-in (slider + tags, optional short note)
  • Trend view over time (weekly / monthly)
  • Privacy-first defaults + clear “not medical advice” wording
  • Crisis support links (shown when needed)

MVP 2 — Education + One Screener + Next Steps

  • Education portal (3–5 core modules) Example: diagnosis meaning, fight/flight/freeze/fawn, what professionals do, basics of stress/anxiety.
  • Weekly screening: start with PHQ-9 + GAD-7 (or DASS-21 if licensing is clear)
  • Next steps: a simple guide + “find help” links (directory later)

This keeps the early build realistic while still supporting the bigger long-term vision.

Safety, ethics, and wording guardrails

These are the basic rules I will follow to keep helloTali safe and responsible:

  • No “you have X” language

    The app will not tell users they have a specific condition (e.g., “you have depression” or “you have dementia”).

  • Not medical advice / not a diagnosis

    helloTali is for education and self-reflection only. It does not provide medical advice or diagnose anyone.

  • Crisis safety message

    If someone is in crisis or in immediate danger, the app will clearly tell them to contact local emergency services or a crisis helpline.

  • Clear consent if any data is stored

    If the app stores any personal data (mood logs, answers, notes), users must clearly opt in and understand what is stored and why.

  • Privacy-first defaults

    The app will collect the minimum data needed. It will prefer storing scores and trends instead of raw sensitive text unless the user chooses otherwise.

Next actions (practical checklist)

1) Pick initial screening instrument

  • Decide: PHQ-9 + GAD-7 vs DASS-21
  • Confirm software usage/licensing for whichever you choose

2) Define the education templates

  • Recommended format for each education page: “What it is / what it isn’t / why it happens / what can help / when to seek help”

3) Write the “Next steps” guide

  • A neutral explainer: GP vs counsellor vs psychologist vs psychiatrist

4) Build one vertical slice (end-to-end preview)

  • (MVP1) Mood check-in → Trend view
  • (MVP2 preview) Education module → (Optional) Screener flow → Next steps

Top comments (0)