Refactor email to Agent Inbox: Resend-only reports with unified agent voice.

Implement Upstash vector integration for semantic job matching and remove Clerk resume vector dependencies.

chore(ui): remove legacy classic launcher paths

chore(ats-direct): enhance discovery + add greenhouse entries

chore(ats-direct): add --fix to registry verifier

chore(ats-direct): add registry verification script

feat(positioning): enforce global-remote guarantee across agent and pipelines

Strengthen global remote intent in prompts and daily pipeline

fix: Groq 429 recovery and 2500ms default score delays

Persist taught job board domains in Redis and merge into digest and pipeline search

Make Groq the default inference provider

Remove Claude/Gemini references

Harden pipeline and email placeholders against undefined access

Add inference failure tracking, billing grace, credits UX, and safer errors

Pipeline: optional MY_CLERK_ID + Clerk resumeVector for hybrid fallback scoring.

Add hybrid MiniLM semantic layer to fallback job scoring.

feat: Search Lab admin experiment, Serper pipeline, heuristic search lab

Remove Gemini/Anthropic; gate WebLLM and server extraction; use OpenRouter for server AI

feat: rolling multi-angle job queries per site (buckets, remote A/B, contract/async)

feat: consolidate Google job queries into OR-groups with URL length packing

Add daily digest fan-out and harden pipeline numeric env parsing.

feat(pipeline): email top 5 close matches when thresholds exclude everything

feat: background batch jobs with Redis, QStash, and launcher precision mode

feat: add Search Lab recency controls and resilient Gemini fallback

feat: migrate pipeline search to SerpAPI with default fallback scoring

fix: avoid OOM by using q4 weights for local model

feat: add local Transformers.js provider + HF Spaces Docker

feat: Open in Job Scorer links in digest + ?url= prefill

fix(ci): use GitHub Environment for pipeline secrets

Add daily job pipeline: CSE discovery, Gemini scoring, email digest