Commit graph

29 commits

Author SHA1 Message Date
Vadym Samoilenko
1b387daacf Migrate task result delivery from WebSocket to HTTP polling
Backend:
- task_manager.py: add result/error/completed_at storage, TTL sweeper (5min), store_task_result() helper
- tasks.py: add GET /<task_id> endpoint returning stored result; cancel route stores 'cancelled' status
- __init__.py: start TTL sweeper on app startup
- All 8 bg functions: store result before emitting lightweight WS hint (no payload data)

Frontend:
- src/lib/taskPolling.ts: waitForTaskResult() — polls GET /tasks/{id} every 2s, WS hint triggers immediate poll, 5min timeout
- src/hooks/useTaskPolling.ts: drop-in replacement for useCancellableGeneration using polling
- Migrate 6 Promise-based WS listeners → waitForTaskResult() in DiscussionPanel, FocusGroupSession (×2), PersonaProfile, PersonaModificationModal, useDiscussionGuideGeneration
- Migrate 3 hook-based consumers → useTaskPolling in AIRecruiter, SyntheticUsers, BulkExportProgressModal

Fixes WS Promise leak: polling survives disconnects, background tabs, page reloads.
WS events retained as zero-payload hints for near-zero latency when connected.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-23 16:46:58 +00:00
Vadym Samoilenko
c7034634e3 Fix all async LLM routes: bypass GCP 30s load balancer timeout
Convert 6 synchronous LLM routes to async 202+WebSocket pattern:
- generate-response (focus_group_ai): persona chat response
- generate-key-themes (focus_group_ai): discussion analysis
- modify-with-ai (personas): AI persona modification
- export-profile (personas): markdown profile export
- describe-asset (focus_groups): image AI description

Each route now returns 202 + task_id immediately, runs LLM in
asyncio background task, delivers result via WebSocket task_completed
event. Frontend listeners updated to wait for ws:task_completed
instead of HTTP response body.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-23 15:56:54 +00:00
Vadym Samoilenko
770bdee829 Remove console.log debug calls from frontend browser console
Removes all debug/verbose console.log calls across frontend to prevent
sensitive data exposure (session IDs, tokens) and reduce console noise.
Keeps only console.error and console.warn for genuine errors.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-23 14:30:56 +00:00
Vadym Samoilenko
2c6505f472 Switch to loginRedirect — popup blocked by server COOP header
Server-level COOP prevents parent from monitoring popup.location,
so loginPopup never resolves in parent. Redirect flow is the correct
solution: page navigates to Microsoft, returns with #code, and
handleRedirectPromise exchanges the idToken with backend.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-20 14:44:19 +00:00
michael
b0445de18b Update GPT-5 to GPT-5.2 and lower default reasoning effort to low
Swap model ID from gpt-5 to gpt-5.2 across all backend services,
frontend components, and documentation. Change default reasoning
effort from medium to low for faster responses.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 11:24:01 -06:00
Michael Clervi
893b537b67 changed permissions 2025-12-19 19:26:16 +00:00
michael
6b7f8763c0 Add environment variable to control local login availability
Add VITE_ENABLE_LOCAL_LOGIN env variable to conditionally show/hide
local username/password login on the login screen. When set to 'false'
(production), only Microsoft login is shown. When 'true' (development),
both options are available.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-08 14:20:50 -06:00
michael
8a9c7c3cb6 Improve discussion guide discoverability in Focus Group session
Add prominent "Edit Discussion Guide" button to header and enhance accordion
styling with amber color scheme, "Click to Edit" badge, and visual feedback
to help users discover the inline editing capability.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-04 09:57:32 -06:00
michael
afa7e43051 Improve draft focus group UX with clickable cards and Run Session button
- Make entire draft card clickable to enter setup mode
- Rename "Edit" button to "Continue Setup" for better clarity
- Add "Run Session" button for complete drafts (with discussion guide and participants)
- Prevent checkbox clicks from triggering card navigation

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-04 09:17:41 -06:00
michael
96b9bfeedd Add LLM-generated one-line summaries for focus groups in list view
- Create focus_group_summary_service.py to generate concise summaries
- Add prompt template for summary generation
- Integrate summary generation after discussion guide creation
- Display summary under focus group title in list view with fallback to description

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-04 08:51:28 -06:00
michael
bbd3f2c828 Add contextual descriptions to AI model selection dropdowns
Users can now see helpful context next to each model option:
- Gemini 3 Pro (Slow, best for most tasks)
- GPT-4.1 (Fast, best for speed)
- GPT-5 (Slow, best for complex tasks)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-01 16:27:57 -06:00
michael
03bb650ec6 Replace inline progress bars with modal progress dialogs for better visibility
- Created new reusable ProgressModal component with animated progress bar
- Converted all inline GenerationProgressBar usages to modal dialogs:
  - AIRecruiter.tsx: Persona generation
  - FocusGroupModerator.tsx: Discussion guide generation
  - FocusGroupSession.tsx: Key themes extraction
  - SyntheticUsers.tsx: Persona summary generation
  - PersonaModificationModal.tsx: Persona modification
- Modal features: auto-dismiss after completion, non-dismissible during operation,
  cancel support, progress animation from 0-90% over 54 seconds
- Fixed broken theme generation state calls in FocusGroupSession.tsx

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-01 11:03:31 -06:00
michael
4d9b0afde7 upgraded to Gemini 3.0 Pro (gemini-3-pro-preview) from Gemini 2.5 Pro
- Upgraded google-genai package from 1.31.0 to 1.52.0
- Updated DEFAULT_MODEL in llm_service.py to gemini-3-pro-preview
- Updated all backend routes, services, and models with new model string
- Updated all frontend components with new model string and display labels
- Updated CLAUDE.md documentation

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-12-01 10:37:36 -06:00
michael
f1cc094318 added full persona profile export in bulk actions menu (CSV, JSON and Markdown formats) 2025-09-10 22:38:35 -05:00
michael
abc9731e4a added hierarchical folders (just two levels) with drag and drop management 2025-09-10 19:53:06 -05:00
michael
e29d2a0bb9 made long actions cancellable (like persona generation, etc.), increased variety of persona generation with prompt changes and temperature variable, reduced length of key theme quotes, bug fixes 2025-09-10 16:24:05 -05:00
michael
8288cb9f5e fixed age to be a single number instead of range, fixed language for AI model to include thematic analysis, and added review/revert/save function to AI persona modification mechanism 2025-09-08 16:10:03 -05:00
michael
1b977ec517 graceful handling of expired JWT token 2025-09-03 13:15:00 -05:00
michael
6a40936508 major refactor of entire application - migrate sync -> async including pymongo -> motor, flask -> quart, google-generativeai -> google-genai 2025-08-27 15:20:56 -05:00
michael
fbef4f42f6 added drag and drop for reordering questions/activities in discussion guide edit mode 2025-08-26 09:52:24 -05:00
michael
3e73259ff3 refactored file uploader and bug fixes 2025-08-13 11:32:56 -05:00
michael
8a5c50cacb refactored image/asset attachment to discussion guide and discussion messages to avoid back end filenames being displayed to user - use metadata system instead. Also added a detailed persona export to markdown. Also, bug fixes 2025-08-12 15:43:34 -05:00
michael
d5dc0bc3af fixed bug related to moderator position status in discussion guide 2025-08-11 16:39:59 -05:00
michael
9a3baa9d97 converted message polling to websockets 2025-08-10 18:08:34 -05:00
michael
fbb444037a fixed folders to be database instead of local storage based, implemented gpt-5, fixed key theme export quotes 2025-08-09 06:38:49 -05:00
michael
3c9518e3ec various bug fixes and UI tweaks 2025-08-07 16:34:37 -05:00
michael
8dcbe7efee various bug fixes and feature additions per Alec's requests 2025-08-06 19:29:27 -05:00
michael
b649793013 added gpt-4.1 support among other things 2025-08-05 17:38:13 -05:00
michael
da7b2c0448 initial commit 2025-08-04 09:07:59 -05:00