Spaces:
Sleeping
Sleeping
Commit History
-
Pre-seed demo candidate interviews so HR dashboard is always populated
husseinelsaadi +1: Claude Opus 4.8The SQLite DB on the Space is ephemeral, so live interviews are wiped on restart, leaving the HR dashboard empty. Seed two completed interviews for the demo candidate (candidate@codingo.ai) on the Data Scientist and Data Engineer jobs (both owned by hr@codingo.ai), with realistic Q&A, evaluations and matching skills. Idempotent per (candidate, job). Now logging into the HR account always shows candidates with working Download Report (PDF) buttons, regardless of restarts. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>981366b -
Fix interview scoring so strong answers score well
husseinelsaadi +1: Claude Opus 4.8Scores were capped at ~50% even for perfect interviews because the final score weighted skills-match (exact string overlap, almost always ~0) at 40%. Rebalance to answer-driven scoring and reduce conservative bias: - interview_api.py / report_generator.py: weight QA answers 85% and skills match 15% (was 40/60); bump category values (Good 0.8->0.85, Medium 0.6->0.65). Mirrored in both so screen + PDF report agree. - interview_engine.py: rewrite the evaluate_answer prompt with a clear rubric, tell LUNA to use the full range and not default to "Medium", and to judge substance not grammar (answers are speech-to-text). Output format unchanged, so parsing is untouched. A flawless interview now scores ~85% (more with matching skills); weak answers still score low. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>c3ec385 -
Seed professional demo data; fix duplicate LUNA avatar on interview
husseinelsaadi +1: Claude Opus 4.8- interview.html: reuse the placeholder avatar/bubble for the first question instead of appending a second LUNA avatar, so only the real question shows. Follow-up questions and talking animation unchanged. - database.py: add idempotent seed_demo_data() (demo recruiter, demo applicant, and 5 curated tech jobs owned by the recruiter) run on startup, so the listing looks polished and the full demo flow works after each ephemeral-DB restart. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>5f99f0f
-
Polish homepage feature cards: per-card accent color, gradient top bar, richer icon tiles
husseinelsaadi +1: Claude Opus 4.8Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>a99ecf6 -
Update homepage to revised design: split hero, deep-indigo navbar, scroll-snap landing
husseinelsaadi +1: Claude Opus 4.8- theme.css: dark branded navbar, two-column hero-split, full-viewport scroll-snap sections scoped to body.landing - index.html: split hero (copy left / LUNA right) + page_class block - base.html: body now carries the page_class block Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>8bfe920 -
Implement Claude Design homepage redesign (preview/index_preview.html)
husseinelsaadi +1: Claude Opus 4.8- theme.css: full premium design system (rotating LUNA halo, float-chip anim, data-reveal, dashboard/tips/application styling) - index.html: hero copy refinements to match the approved preview - base.html: additive scroll-reveal IntersectionObserver Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>0334723 -
Implement Codingo SaaS hero redesign on landing page
husseinelsaadi +1: Claude Opus 4.8Add hero-stage with floating live-status chips and a gradient LUNA wordmark, ported from the Claude Design "Codingo SaaS redesign" preview. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>905dc3f -
Richer CV experience, sharper STT, livelier LUNA voice
husseinelsaadi +1: Claude Opus 4.8- CV parsing now captures each role's responsibilities/description (not just title+company); experience/education are split on newlines so descriptions with commas stay intact - Interview prompts direct LUNA to ask specific questions about the candidate's actual projects and technologies - Speech-to-text: default to the English small.en Whisper model and bias decoding with a technical-vocabulary initial_prompt (fixes "roles"->"Rolls") - LUNA's voice now speaks ~15% faster (tunable via LUNA_TTS_RATE) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>c5a47ef -
Switch UI to a clean, professional light theme
husseinelsaadi +1: Claude Opus 4.8- Light glass navbar, ink logo/links, light-surface footer - Replace the dark cinematic hero with a near-white premium hero (restrained brand-tint wash, ink text, soft restrained LUNA glow) - Buttons restyled for light backgrounds (visible outline/logout buttons) - Lighten the interview page background and the closing score card to match - Brand gradient kept only as a sparing accent (logo, primary buttons, CTA band) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>4fb152a -
Accurate LLM CV parsing + realistic, memory-aware interview
husseinelsaadi +1: Claude Opus 4.8CV parsing (resume_parser.py, app.py): - Parse resumes with the Groq LLM (clean name/skills/education/experience JSON), far more accurate than the old keyword/regex heuristics; keep the regex/NER path as a lazy fallback so startup no longer loads the heavy NER model - Auto-parse the uploaded CV on apply and populate the profile (manual form fields still override), so an ATS-friendly CV "just works" Interview realism (interview_engine.py, interview_api.py): - Give follow-ups real memory: pass the persisted interview_log as conversation_history (the frontend never sent it) - Evaluate answers at the real job role + seniority instead of a generic default - Enrich prompts with job skills/seniority/description; LUNA now acknowledges the last answer and asks one focused, contextual follow-up - Fix an empty-experience IndexError that silently dropped the first question Also includes earlier deploy hardening: env-driven SECRET_KEY/FLASK_DEBUG and removal of a duplicate generate_next…c28f81f -
Premium UI redesign, repo cleanup, and restored LUNA media
husseinelsaadi +1: Claude Opus 4.8- Add theme.css design system (brand tokens, Inter/Space Grotesk, glass navbar, gradient buttons, premium cards) linked across all pages - Cinematic dark hero + animated "How it works" pipeline on the landing page - Rebrand the interview screen as "Interview with LUNA": her real photo as the avatar, speaking equalizer, brand-colored bubbles/buttons (no JS changes) - Polished closing score card and multi-column footer - Restore the real LUNA.png + demo video (were broken Git-LFS stubs) and stop tracking png/mp4 via LFS so the Space always has the real binaries - Ignore the local archive/ folder; drop the unused data/merged_dataset.json - Add docs/social-video-guide.md (LUNA explainer-video production guide) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>b5d621a
-
Fix LUNA voice + transcription: edge-tts 7.2.8 (fixes 403), Whisper base model + en/VAD for accuracy
husseinelsaadi +1: Claude Opus 4.8Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>145f34b -
Revive Codingo: free-CPU deploy, env-driven Qdrant, SQLite fallback, CPU crash fix
husseinelsaadi +1: Claude Opus 4.8- Dockerfile: CUDA base -> python:3.10-slim CPU; install CPU torch; add spaCy model - interview_retrieval.py: read Qdrant URL/key from env instead of dead hardcoded cluster - interview_engine.py: guard unconditional GPU calls that crashed on CPU at import - app.py: fall back to SQLite when DATABASE_URL is unset; normalise postgres:// scheme - requirements.txt: drop GPU-only bitsandbytes; add nltk + python-dateutil for resume parser - add scripts/rebuild_qdrant.py + data/ to rebuild the interview-question vector DB Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>41604f6
-
updated
e0470f3 -
updated
88df9ba -
updated
14500de -
updated
8330d5a -
check up done
cb47676
-
updated
aea35a7
-
updated
b822cf0 -
updated
194e7a7 -
updated
aba3be2 -
updated report
06654d0 -
updated
d32e6aa -
updated
45f0a42 -
upadted
471f933 -
updated
b8deff5 -
resume parser updated
288175b
-
updated
c0dac84 -
updated
f2a1cfa -
updated
6248af7 -
updated
2489359 -
updated
2c6f29c -
updated
7ec2f5a -
updated
33fa314 -
updated
682910e -
updated
947d727 -
updated
6d286f1 -
updated
a511250 -
updated
864c2ae -
updated
efffc2e -
updated
f3f24e3 -
updated
4f1e97d -
updated
8d99522 -
updated
1ead253 -
gemini updated
0e43f07 -
updated
722e882 -
updated
2e9d967 -
updated
1de8504 -
switched to falcon
2a8ccfd