Skip to content

Latest commit

 

History

History
365 lines (296 loc) · 11.8 KB

hackathon_github_issues.md

File metadata and controls

365 lines (296 loc) · 11.8 KB

Hackathon MVP — GitHub Issues (24‑Hour Prototype)

Context: Issues derived from the AI‑Powered Student Success Analytics PRD.
Use: This file is machine‑parsable by Cursor and easy to convert into gh issue create commands.
Convention: Each issue block is delimited by --- and includes keys: Title, Labels, Estimate, and Body (GitHub‑flavored Markdown).

Priority levels: P0 (Must‑Have), P1 (Nice‑to‑Have), P2 (Stretch)

Quick Start (optional)

You can script creation like:

# Example: create first issue block manually
gh issue create   --title "MVP data schema & seed AR/PDP sample"   --body "$(awk '/^---/{f++} f==1{print}' hackathon_github_issues.md | sed '1,4d')"   --label "priority:P0" --label "area:data" --label "type:feature" --label "hackathon"

(Or let Cursor generate a script to iterate over blocks.)

Title: MVP data schema & seed AR/PDP sample in Supabase Labels: ["priority:P0","area:data","type:feature","hackathon","mvp"] Estimate: 2h Body:

Summary

Define a minimal relational schema in Supabase to hold PDP/AR-like data (students, terms, courses, enrollments, outcomes/DFWI). Seed with a small synthetic sample for each institution.

Why

Unblocks every downstream feature (dashboards, NLQ, baseline ML). Supabase provides instant REST/GraphQL APIs.

Acceptance Criteria

  • Tables created in Supabase for: students, terms, courses, enrollments, outcomes_dfwi.
  • Sample rows loaded for KCTCS, Bishop State, Akron.
  • README shows Supabase setup & connection details.

Tasks

  • Create Supabase project and SQL migrations.
  • Write a loader script to insert sample CSVs.
  • Provide ERD diagram (exported PNG).

Out of Scope

Full PDP spec parity.

Title: ETL loader for CSV (local) → Supabase Labels: ["priority:P0","area:data","type:feature","hackathon","etl"] Estimate: 2h Body:

Summary

Minimal Python ETL script that loads local CSV into Supabase from /data/*.csv using Supabase client library.

Why

Consistent repeatable load to refresh the prototype quickly during the hackathon.

Acceptance Criteria

  • make load or python scripts/load_csv.py ingests all CSVs to Supabase.
  • Logs rows upserted per table.
  • Idempotent (can re-run without dupes).

Tasks

  • Implement CSV reader + UPSERTs using Supabase Python client.
  • Store Supabase credentials in .env.
  • Document in README.

Title: PDP pre‑submission validator (lightweight) Labels: ["priority:P0","area:data","type:tooling","hackathon"] Estimate: 1.5h Body:

Summary

Python validator that checks common PDP formatting issues (required columns present, types OK, no nulls in keys).

Why

Addresses PRD callout for reducing latency from PDP rejections.

Acceptance Criteria

  • CLI: python tools/pdp_validate.py path/to/file.csv -> clear pass/fail + line counts.
  • Unit tests for 3 common error cases.

Tasks

  • Define minimal rules (YAML) and validator.
  • Add sample bad file in /fixtures.

Title: Supabase API setup & custom Edge Functions Labels: ["priority:P0","area:backend","type:feature","hackathon","api"] Estimate: 2h Body:

Summary

Configure Supabase auto-generated REST API and create custom Edge Functions for complex queries:

  • /metrics/retention?term=...
  • /metrics/gateway_completion?course_id=...
  • /students/at_risk?limit=...

Why

Supabase provides instant CRUD APIs; Edge Functions handle custom business logic for dashboard and NLQ layer.

Acceptance Criteria

  • Supabase REST API configured with RLS policies.
  • Custom Edge Functions deployed for complex metrics.
  • Returns JSON from Supabase queries.

Tasks

  • Configure Supabase API settings & auth.
  • Implement Edge Functions for custom metrics.
  • Test API endpoints with Postman/curl.

Title: React dashboard (MVP) with 3 core widgets Labels: ["priority:P0","area:frontend","type:feature","hackathon","ui"] Estimate: 3h Body:

Summary

A simple React (Vite) dashboard that calls the Supabase API and shows:

  1. Retention % by cohort/term
  2. Gateway course completion %
  3. DFWI by course

Why

Visual proof of value for judges; matches PRD core KPIs.

Acceptance Criteria

  • Local dev: npm run dev works.
  • Charts render from Supabase API, not hard-coded.
  • Institution switcher (dropdown) filters data.
  • Supabase JS client integrated.

Tasks

  • Set up project with Tailwind + Supabase client.
  • Charting lib (Recharts).
  • Minimal routing + error states.

Title: NLQ stub: natural‑language → SQL (rule‑based) Labels: ["priority:P0","area:ai","type:feature","hackathon","nlq"] Estimate: 2h Body:

Summary

Implement a rule‑based mapper (Supabase Edge Function or Python service) for whitelisted queries (e.g., "retention last term", "DFWI for COURSE101"). Avoids full LLM dependency for MVP.

Why

Demonstrates AI‑assisted experience in 24h.

Acceptance Criteria

  • Edge Function or endpoint /nlq?q=... returns JSON with the matched metric + underlying query used.
  • Support 5–7 canonical phrasings.
  • Leverages Supabase queries for data fetching.

Tasks

  • Define patterns → Supabase query templates.
  • Reuse existing metrics queries.

Out of Scope

Free‑form arbitrary SQL generation.

Title: Baseline at‑risk score (heuristic) Labels: ["priority:P0","area:ml","type:feature","hackathon","mvp"] Estimate: 2h Body:

Summary

Compute a naive "at‑risk" score per student using simple features: prior D/F/W/I counts, gateway completion, credits completed.

Why

Delivers predictive flavor without long model training.

Acceptance Criteria

  • Supabase view or Edge Function for /students/at_risk returns top N with score & reasons.
  • Document formula in README.

Tasks

  • Create Supabase SQL view or stored function to compute score.
  • Expose via Edge Function or direct RPC call.

Title: Weekly refresh job (cron) — simulated Labels: ["priority:P0","area:platform","type:chore","hackathon","ops"] Estimate: 1h Body:

Summary

Add a containerized cron (or watch) that re-runs ETL loader on a schedule (simulate “weekly” as every 5 minutes in dev).

Why

Matches PRD “weekly refresh” expectation.

Acceptance Criteria

  • Job logs visible in docker compose logs.
  • Re-loads CSVs without downtime.

Tasks

  • Add lightweight scheduler container.
  • Document how to toggle frequency.

Title: Supabase Auth setup (optional for MVP) Labels: ["priority:P1","area:security","type:feature","hackathon"] Estimate: 1.5h Body:

Summary

Set up basic Supabase authentication with email/password or magic link for demo purposes.

Why

Demonstrates security best practices and leverages Supabase's built-in auth.

Acceptance Criteria

  • Supabase Auth configured with at least one provider.
  • Login/logout flow in React UI.
  • RLS policies protect sensitive data.

Tasks

  • Enable Supabase Auth in project settings.
  • Implement login UI component.
  • Configure basic RLS policies.

Title: Role‑based views (very light) Labels: ["priority:P0","area:security","type:feature","hackathon","mvp"] Estimate: 1h Body:

Summary

UI visibility toggles for 2 roles: Advisor (show at‑risk list) and Leadership (show ROI/retention charts). Leverage Supabase Auth user metadata for roles or use .env ROLE=advisor|leadership for MVP.

Why

Demonstrates accessibility angle with minimal IAM setup.

Acceptance Criteria

  • Role-based widget visibility (from Supabase Auth or .env).
  • Document behavior.

Title: FERPA/PII hygiene pass (MVP) Labels: ["priority:P0","area:compliance","type:chore","hackathon"] Estimate: 1h Body:

Summary

Ensure sample data is synthetic. Mask or suppress direct identifiers in UI/API by default (use internal IDs).

Why

Aligns with compliance best practices.

Acceptance Criteria

  • No student names/SSNs exposed.
  • UI shows anonymized IDs only.
  • README lists data provenance.

Title: Demo script & screenshots Labels: ["priority:P0","area:docs","type:documentation","hackathon"] Estimate: 1h Body:

Summary

Short script for a 5–7 minute demo + 4 screenshots of key widgets.

Why

Ensures a tight, rehearsed final presentation.

Acceptance Criteria

  • DEMO.md with talk track + commands.
  • /shots/*.png added.

Title: One‑page ROI summary per institution Labels: ["priority:P1","area:docs","type:documentation","hackathon"] Estimate: 1h Body:

Summary

Brief write‑ups mapping dashboard value to each institution (KCTCS, Bishop State, Akron) and their use cases.

Acceptance Criteria

  • docs/roi_kctcs.md, docs/roi_bishop_state.md, docs/roi_akron.md

Title: Course sequence insight (DFWI by combo) — static query Labels: ["priority:P1","area:data","type:feature","hackathon"] Estimate: 1.5h Body:

Summary

Provide a precomputed Supabase view for DFWI by common course combinations (top 10 combos) and expose via API.

Acceptance Criteria

  • Supabase SQL view + accessible via REST API + UI table widget.

Title: Email export (CSV) of at‑risk list Labels: ["priority:P1","area:frontend","type:feature","hackathon","export"] Estimate: 1h Body:

Summary

Button on Advisor view to export current at‑risk table as CSV.

Acceptance Criteria

  • Downloads CSV reflecting current filters.

Title: Health & diagnostics page Labels: ["priority:P1","area:platform","type:chore","hackathon","ops"] Estimate: 1h Body:

Summary

Small UI page that checks Supabase connection status and displays last ETL refresh timestamp.

Acceptance Criteria

  • Green/Red indicators with timestamps.
  • Shows Supabase API connectivity status.

Title: LLM‑backed NLQ (behind a flag) Labels: ["priority:P2","area:ai","type:feature","hackathon","nlq"] Estimate: 3h Body:

Summary

Swap the rule‑based NLQ with an LLM call if ENABLE_LLM=1. Keep same query whitelist.

Acceptance Criteria

  • Feature flag respected.
  • Logs prompt + chosen SQL safely.

Title: Simple retention baseline model (logistic regression) Labels: ["priority:P2","area:ml","type:feature","hackathon"] Estimate: 3h Body:

Summary

Train a fast sklearn logistic regression on sample historical features from Supabase to classify "retained next term" (for demo).

Acceptance Criteria

  • Training script with saved coefficients.
  • Edge Function or batch job to compute predictions and store back in Supabase.

Title: PowerBI/Embedding spike (notes only) Labels: ["priority:P2","area:integration","type:spike","hackathon"] Estimate: 1h Body:

Summary

Investigate feasibility of embedding Supabase API data into PowerBI/Looker equivalents; capture notes and blockers for post‑hackathon.

Acceptance Criteria

  • spikes/powerbi_embed.md committed.
  • Document Supabase REST API → PowerBI connector options.

Title: Packaging & one‑command bootstrap Labels: ["priority:P1","area:devex","type:chore","hackathon"] Estimate: 1h Body:

Summary

Single command (make dev or ./bootstrap.sh) to set up environment and start the UI (Supabase is cloud-hosted).

Acceptance Criteria

  • Works on clean machine with Node + Python preinstalled.
  • Configures .env with Supabase credentials.
  • README updated.

Title: README (root) with setup & architecture Labels: ["priority:P0","area:docs","type:documentation","hackathon","mvp"] Estimate: 1h Body:

Summary

Clear, concise README: architecture diagram showing Supabase integration, services, how to run, and demo flow.

Acceptance Criteria

  • Architecture diagram image (showing Supabase as backend/DB).
  • Copy‑paste commands for local run.
  • Supabase setup instructions.
  • Link to DEMO.md.