- Dockerfile.finetune: PyTorch 2.3/CUDA 12.1 base + unsloth + training stack
- finetune_local.py: auto-register model via Ollama HTTP API after GGUF
export; path-translate between finetune container mount and Ollama's view;
update config/llm.yaml automatically; DOCS_DIR env override for Docker
- prepare_training_data.py: DOCS_DIR env override so make prepare-training
works correctly inside the app container
- compose.yml: add finetune service (cpu/single-gpu/dual-gpu profiles);
DOCS_DIR=/docs injected into app + finetune containers
- compose.podman-gpu.yml: CDI device override for finetune service
- Makefile: make prepare-training + make finetune targets
Interactive prompt lets users with split-drive setups point Ollama and
vLLM model dirs at a dedicated storage drive. Reads current .env value
as default so re-runs are idempotent. Skips prompts in non-interactive
(piped) mode. Creates the target directory immediately and updates .env
in-place via portable awk (Linux + macOS). Also simplifies next-steps
output since model paths are now configured at install time.
- llm.yaml + example: replace localhost URLs with Docker service names
(ollama:11434, vllm:8000, vision:8002); replace personal model names
(alex-cover-writer, llama3.1:8b) with llama3.2:3b
- user.yaml.example: update service hosts to Docker names (ollama, vllm,
searxng) and searxng port from 8888 (host-mapped) to 8080 (internal)
- wizard step 5: fix hardcoded localhost defaults — wizard runs inside
Docker, so service name defaults are required for connection tests to pass
- scrapers/companyScraper.py: bundle scraper so Dockerfile COPY succeeds
- setup.sh: remove host Ollama install (conflicts with Docker Ollama on
port 11434); Docker entrypoint handles model download automatically
- README + setup.sh banner: add Circuit Forge mission statement
- generate() accepts previous_result + feedback; appends both to LLM prompt
- task_runner cover_letter handler parses params JSON, passes fields through
- Apply Workspace: "Refine with Feedback" expander with text area + Regenerate
button; only shown when a draft exists; clears feedback after submitting
- 8 new tests (TestGenerateRefinement + TestTaskRunnerCoverLetterParams)
Replaces the old 5-step wizard with a 7-step orchestrator that uses the
step modules built in Tasks 2-8. Steps 1-6 are mandatory (hardware, tier,
identity, resume, inference, search); step 7 (integrations) is optional.
Each Next click validates, writes wizard_step to user.yaml for crash recovery,
and resumes at the correct step on page reload. LLM generation buttons
submit wizard_generate tasks and poll via @st.fragment(run_every=3). Finish
sets wizard_complete=True, removes wizard_step, and calls apply_service_urls.
Adds tests/test_wizard_flow.py (7 tests) covering validate() chain, yaml
persistence helpers, and wizard state inference.
Add all 13 integration modules (Notion, Google Drive, Google Sheets,
Airtable, Dropbox, OneDrive, MEGA, Nextcloud, Google Calendar, Apple
Calendar/CalDAV, Slack, Discord, Home Assistant) with fields(), connect(),
and test() implementations. Add config/integrations/*.yaml.example files
and gitignore rules for live config files. Add 5 new registry/schema
tests bringing total to 193 passing.
- Add tier, dev_tier_override, wizard_complete, wizard_step, dismissed_banners
fields to UserProfile with defaults and effective_tier property
- Add params TEXT column to background_tasks table (CREATE + migration)
- Update insert_task() to accept params with params-aware dedup logic
- Update submit_task() and _run_task() to thread params through
- Add test_wizard_defaults, test_effective_tier_override,
test_effective_tier_no_override, and test_insert_task_with_params
scripts/preflight.py (stdlib-only, no psutil):
- Port probing: owned services auto-reassign to next free port; external
services (Ollama) show ✓ reachable / ⚠ not responding
- System resources: CPU cores, RAM (total + available), GPU VRAM via
nvidia-smi; works on Linux + macOS
- Profile recommendation: remote / cpu / single-gpu / dual-gpu
- vLLM KV cache offload: calculates CPU_OFFLOAD_GB when VRAM < 10 GB
free and RAM headroom > 4 GB (uses up to 25% of available headroom)
- Writes resolved values to .env for docker compose; single-service mode
(--service streamlit) for scripted port queries
- Exit 0 unless an owned port genuinely can't be resolved
scripts/manage-ui.sh:
- Calls preflight.py --service streamlit before bind; falls back to
pure-bash port scan if Python/yaml unavailable
compose.yml:
- vllm command: adds --cpu-offload-gb ${CPU_OFFLOAD_GB:-0}
Makefile:
- start / restart depend on preflight target
- PYTHON variable for env portability
- test target uses PYTHON variable
scripts/migrate.py:
- dry-run by default; --apply writes files; --copy-db migrates staging.db
- generates config/user.yaml from source repo's resume + cover letter scripts
- copies gitignored configs (notion, email, adzuna, craigslist, search profiles,
resume keywords, blocklist, aihawk resume)
- merges fine-tuned model name from source llm.yaml into dest llm.yaml
scripts/manage-ui.sh:
- STREAMLIT_BIN no longer hardcoded; auto-resolves via conda env or PATH;
override with STREAMLIT_BIN env var
scripts/manage-vllm.sh:
- VLLM_BIN and MODEL_DIR now read from env vars with portable defaults
LGBTQIA+ inclusion section in research briefs:
- user_profile.py: add candidate_lgbtq_focus bool accessor
- user.yaml.example: add candidate_lgbtq_focus flag (default false)
- company_research.py: gate new LGBTQIA+ section behind flag; section
count now dynamic (7 base + 1 per opt-in section, max 9)
- 2_Settings.py: add "Research Brief Preferences" expander with
checkboxes for both accessibility and LGBTQIA+ focus flags;
mission_preferences now round-trips through save (no silent drop)
Phase 2 fixes:
- manage-vllm.sh: MODEL_DIR and VLLM_BIN now read from env vars
(VLLM_MODELS_DIR, VLLM_BIN) with portable defaults
- search_profiles.yaml: replace personal CS/TAM/Bay Area profiles
with a documented generic starter profile
Phase 3 fix:
- llm.yaml: rename alex-cover-writer:latest → llama3.2:3b with
inline comment for users to substitute their fine-tuned model;
fix model-exclusion comment
- 3_Resume_Editor.py: replace "Alex's" in docstring and caption
- user_profile.py: expose mission_preferences and candidate_accessibility_focus
- user.yaml.example: add mission_preferences section + candidate_accessibility_focus flag
- generate_cover_letter.py: build _MISSION_NOTES from user profile instead of
hardcoded personal passion notes; falls back to generic defaults when not set
- company_research.py: gate "Inclusion & Accessibility" section behind
candidate_accessibility_focus flag; section count adjusts (7 or 8) accordingly
- UserProfile class drives all personal data
- First-run wizard gates app until user.yaml exists
- Docker Compose stack: remote/cpu/single-gpu/dual-gpu profiles
- Vision service containerized (single-gpu/dual-gpu)
- All Alex/Library references removed from app and scripts
- Circuit Forge LLC / Peregrine branding throughout
- Add moondream2 vision service to compose.yml (single-gpu + dual-gpu profiles)
- Create scripts/vision_service/Dockerfile for the vision container
- Add VISION_PORT, VISION_MODEL, VISION_REVISION vars to .env.example
- Add Vision Service entry to SERVICES list in Settings (hidden unless gpu profile active)
- Add Fine-Tune Wizard tab (Task 10) to Settings with 3-step upload→preview→train flow
- Tab is always rendered; shows info message when non-GPU profile is active
Replace hardcoded systemd/shell-script service commands with docker compose
profile-aware commands. Add inference_profile-based filtering (hidden flag
removes Ollama on remote profile, vLLM unless dual-gpu). Replace TCP socket
health check with HTTP-based _port_open() that accepts host/ssl/verify params
for remote/TLS-terminated service support.
