SemiAnalysisAI · adibarra · Jul 27, 2026 · Jul 27, 2026 · Jul 27, 2026
@@ -23,7 +23,7 @@ You ingest benchmark runs from `SemiAnalysisAI/InferenceX` GitHub Actions into t
 cd /Users/quilicic/InferenceX-app/packages/db
 DATABASE_WRITE_URL='<provided direct non-pooled write URL>' \
 GITHUB_TOKEN=$(gh auth token) \
-pnpm exec tsx src/ingest-ci-run.ts --download <RUN_ID> SemiAnalysisAI/InferenceX
+bun src/ingest-ci-run.ts --download <RUN_ID> SemiAnalysisAI/InferenceX
 ```
 
 Then refresh the materialized view (the script's auto-refresh sometimes races):
@@ -118,7 +118,7 @@ cd /Users/quilicic/InferenceX-app/packages/db
 INGEST_RUN_ID=$RID INGEST_RUN_ATTEMPT=1 INGEST_ARTIFACTS_PATH=$TMPDIR INGEST_REPO=SemiAnalysisAI/InferenceX \
 DATABASE_WRITE_URL='<provided direct non-pooled write URL>' \
 GITHUB_TOKEN=$(gh auth token) \
-pnpm exec tsx src/ingest-ci-run.ts
+bun src/ingest-ci-run.ts
 rm -rf $TMPDIR
 ```
 
@@ -162,7 +162,7 @@ If the user doesn't specify a description, DO NOT skip the entry and DO NOT bloc
 - **Multi-attempt artifacts**: a single GitHub run can spill across runners (`h200-cw_00` + `h200-dgxc-slurm_1`); the logical-name dedup strips the `_<runner>_<attempt>` suffix.
 - **Materialized view dedup tiebreaker**: `latest_benchmarks` picks rows by `date DESC, wr.run_started_at DESC`. Backfilling old data may not surface unless dates align with the user's date picker selection.
 - **Date alignment for partial runs**: when a re-run only covers a subset of concs (`replace ONLY the points this run produces`), align dates with prior full sweep via `UPDATE benchmark_results.date = '<full-sweep-date>'` so the frontend's max-date-per-group dedup doesn't drop the older sweep.
-- **Agentic interactivity normalization (`*_intvty`)**: for `agentic_traces` runs, interactivity MUST be the slow-tail reciprocal of the ITL percentile — `*_intvty = 1/*_itl` (so `p90_intvty = 1/p90_itl`). Some harness versions emit `*_intvty` as `p(1/ITL)` instead (fast-tail — inverts percentile order, e.g. p90 shows ~`1/p10(ITL)`), which silently contaminates cross-run Pareto comparisons. The ingest mapper (`benchmark-mapper.ts`) now **derives `*_intvty` from `*_itl` and discards the artifact's value** for agentic rows, so a normal ingest is self-correcting — no manual step needed. The frontend `agenticAliases` does the same for overlay / `?unofficialrun=` rows. If you ever load agentic data through a path that bypasses the mapper, run `pnpm --filter @semianalysisai/inferencex-db db:backfill-agentic-intvty --yes` (idempotent; rewrites `mean/p75/p90/p95 _intvty = 1/_itl`) then refresh the MV + purge cache. `std_intvty` is intentionally left alone (the reciprocal of a std is meaningless; the API strips it anyway).
+- **Agentic interactivity normalization (`*_intvty`)**: for `agentic_traces` runs, interactivity MUST be the slow-tail reciprocal of the ITL percentile, so `p90_intvty = 1/p90_itl`. Some harness versions emit `p(1/ITL)` instead, which inverts percentile order and contaminates cross-run Pareto comparisons. The ingest mapper derives `*_intvty` from `*_itl` and discards the artifact value for agentic rows. The frontend `agenticAliases` does the same for overlay and `?unofficialrun=` rows. Do not ingest through a path that bypasses these normalizers; the retired one-shot backfill is no longer available. `std_intvty` stays unchanged because the reciprocal of a standard deviation is meaningless, and the API strips it.
 
 ## Process
 
@@ -180,7 +180,7 @@ This agent ingests **benchmark runs**. The HF agentic trace **datasets** (`semia
 
 ```bash
 cd packages/db && DATABASE_WRITE_URL='<direct write url>' \
-  pnpm exec tsx src/ingest-weka-dataset.ts <hf-dataset-id> \
+  bun src/ingest-weka-dataset.ts <hf-dataset-id> \
   [--label "…"] [--variant full|256k] [--description "…"] [--limit N]
 ```
 

@@ -1,5 +1,5 @@
 ---
-allowed-tools: Bash(git log:*), Bash(git diff:*), Bash(git blame:*), Bash(pnpm test:*), Bash(pnpm typecheck*), Bash(pnpm lint*), Bash(pnpm dev*), Bash(curl:*), Read, Glob, Grep
+allowed-tools: Bash(git log:*), Bash(git diff:*), Bash(git blame:*), Bash(bun run test:*), Bash(bun run typecheck*), Bash(bun run lint*), Bash(bun run dev*), Bash(curl:*), Read, Glob, Grep
 description: Systematic debugging — root cause before fixes
 ---
 
@@ -57,7 +57,7 @@ BEFORE attempting ANY fix:
 
 1. **Create Failing Test** — regression test reproducing the bug with exact triggering input (per CLAUDE.md testing requirements)
 2. **Implement Single Fix** — address root cause, ONE change, no "while I'm here" improvements
-3. **Verify Fix** — run `pnpm test:unit` and `pnpm typecheck`, confirm no other tests broken
+3. **Verify Fix** — run `bun run test:unit` and `bun run typecheck`, confirm no other tests broken
 
 ## Red Flags — STOP and Return to Phase 1
 

@@ -1,5 +1,5 @@
 ---
-allowed-tools: Bash(pnpm build*), Bash(pnpm typecheck*), Bash(pnpm lint*), Bash(pnpm fmt*), Read, Edit, Glob, Grep
+allowed-tools: Bash(bun run build*), Bash(bun run typecheck*), Bash(bun run lint*), Bash(bun run fmt*), Read, Edit, Glob, Grep
 description: Incrementally fix build, type, and lint errors with minimal safe changes
 ---
 
@@ -12,14 +12,14 @@ Incrementally fix build and type errors with minimal, safe changes.
 Run all checks and capture errors:
 
 ```bash
-pnpm typecheck 2>&1
-pnpm lint 2>&1
+bun run typecheck 2>&1
+bun run lint 2>&1
 ```
 
 If both pass, run the full build:
 
 ```bash
-pnpm build 2>&1
+bun run build 2>&1
 ```
 
 If everything passes, announce "All checks pass — nothing to fix." and stop.
@@ -47,7 +47,7 @@ For each error:
 - A fix introduces **more errors than it resolves**
 - The **same error persists after 3 attempts**
 - The fix requires **architectural changes** or touching >3 files
-- Errors stem from **missing dependencies** (need `pnpm install`)
+- Errors stem from **missing dependencies** (need `bun install`)
 
 ## Step 5: Summary
 

@@ -1,5 +1,5 @@
 ---
-allowed-tools: Bash(pnpm test:*), Bash(pnpm typecheck*), Bash(pnpm lint*), Bash(pnpm build*), Bash(pnpm dev*), Bash(pnpm fmt*), Bash(curl:*), Bash(git diff:*), Bash(git status*), Read, Glob, Grep
+allowed-tools: Bash(bun run test:*), Bash(bun run typecheck*), Bash(bun run lint*), Bash(bun run build*), Bash(bun run dev*), Bash(bun run fmt*), Bash(curl:*), Bash(git diff:*), Bash(git status*), Read, Glob, Grep
 description: Verify work is complete before committing — evidence before claims
 ---
 
@@ -34,39 +34,39 @@ Run each step. Report the actual output — not what you expect.
 ### 1. Type checking
 
 ```bash
-pnpm typecheck
+bun run typecheck
 ```
 
 Required: exit 0, no errors
 
 ### 2. Linting
 
 ```bash
-pnpm lint
+bun run lint
 ```
 
 Required: exit 0, no errors
 
 ### 3. Formatting
 
 ```bash
-pnpm fmt
+bun run fmt
 ```
 
 Required: exit 0, no formatting issues
 
 ### 4. Unit tests
 
 ```bash
-pnpm test:unit
+bun run test:unit
 ```
 
 Required: all tests pass, 0 failures
 
 ### 5. Dev server starts
 
 ```bash
-pnpm dev --hostname 0.0.0.0 --port 3000 &
+bun run dev -- --hostname 0.0.0.0 --port 3000 &
 curl --retry 10 --retry-delay 2 --retry-connrefused -sSf http://localhost:3000 >/dev/null
 ```
 
@@ -75,7 +75,7 @@ Required: server responds successfully
 ### 6. E2E tests
 
 ```bash
-pnpm test:e2e
+bun run test:e2e
 ```
 
 Required: all tests pass

@@ -58,14 +58,14 @@ Use this format:
       [Exact test code]
 
 - [ ] **Step 2: Run test to verify it fails**
-      Run: `pnpm test:unit -- path/to/test`
+      Run: `bun run test:unit -- path/to/test`
       Expected: FAIL
 
 - [ ] **Step 3: Write minimal implementation**
       [Exact implementation code]
 
 - [ ] **Step 4: Run test to verify it passes**
-      Run: `pnpm test:unit -- path/to/test`
+      Run: `bun run test:unit -- path/to/test`
       Expected: PASS
 
 - [ ] **Step 5: Commit**

@@ -12,7 +12,7 @@
         "hooks": [
           {
             "type": "command",
-            "command": "pnpm fmt:fix; pnpm lint:fix; true",
+            "command": "bun run fmt:fix; bun run lint:fix; true",
             "timeout": 10,
             "async": true
           }

@@ -321,7 +321,7 @@ The editor runs as a background Node process on `127.0.0.1:4747`, reads from and
 
 While the user reviews in the browser, you can:
 
-- Run `pnpm lint && pnpm typecheck` against the working tree to catch any MDX errors that would block the pre-commit hook later.
+- Run `bun run lint && bun run typecheck` against the working tree to catch any MDX errors that would block the pre-commit hook later.
 - Save the chart image into `packages/app/public/images/{slug}/benchmark-light.png` (and `benchmark-dark.png` if the user provided both) so the `<Figure>` placeholder in the preview shows a real path.
 
 **Concurrent-edit collision warning.** The browser editor auto-saves the user's textarea ~800 ms after their last keystroke. If you re-edit a paragraph the user has open in CodeMirror, your `Edit` call writes to disk first, then the editor's debounced save overwrites your change with the user's stale buffer the next time they type or the timer fires. Failure mode: user asks you to expand a paragraph, you expand it on disk, user types one more character in the browser, the one-liner comes back. When you need to edit a section the user is actively working on, **tell the user explicitly to either close the browser tab or hit the "↻ Reload from disk" button before resuming editing**. Don't rely on them noticing the collision — it looks like nothing happened from their side.
@@ -342,7 +342,7 @@ git push -u origin blog/{slug}
 gh pr create --title "feat(blog): ..." --body "..."
 ```
 
-The pre-commit hook runs `oxlint`, `oxfmt`, and `tsc --noEmit`. All three must pass. If lint/format fails, run `pnpm lint:fix && pnpm fmt:fix` and re-commit (don't `--no-verify`).
+The pre-commit hook runs `oxlint`, `oxfmt`, and `tsc --noEmit`. All three must pass. If lint/format fails, run `bun run lint:fix && bun run fmt:fix` and re-commit (don't `--no-verify`).
 
 After the PR opens, expect Cursor Bugbot to flag correctness issues in the prose (numeric overstatement, claims contradicted by tables, wrong attribution). Treat its findings as real review comments — fix them in a follow-up commit, then resolve the threads. Branch protection on master requires resolved review threads before auto-merge fires.
 

@@ -18,12 +18,12 @@ This prompt is runtime context. Do not assume any file path from it without veri
   - If you touched inference or evaluation, verify the unofficial-run overlay path too. AGENTS.md §"Unofficial Run Support" lists the exact code paths and the `?unofficialrun=<github-actions-run-id>` URL pattern; do not skip it.
   - New interactive elements get a `track()` call per AGENTS.md §"Analytics Requirement" (`[section]_[action]` naming).
   - Tests added/updated per `docs/testing.md` (unit tests colocated as `<module>.test.ts`; E2E tests in `packages/app/cypress/e2e/`). Missing or low-quality tests are blocking.
-  - Run `pnpm typecheck && pnpm lint && pnpm test:unit && pnpm test:e2e` — all green before commit.
+  - Run `bun run typecheck && bun run lint && bun run test:unit && bun run test:e2e` — all green before commit.
 - code — backend / DB / ETL / ingest / lib / API-route logic, no UI surface:
-  - Add or update colocated unit tests (`<module>.test.ts`) per `docs/testing.md`; run `pnpm typecheck && pnpm lint && pnpm test:unit`.
+  - Add or update colocated unit tests (`<module>.test.ts`) per `docs/testing.md`; run `bun run typecheck && bun run lint && bun run test:unit`.
   - Only if your change turns out to be user-visible: self-provision a browser (see "Browser verification"), then apply the `ui` rules above.
 - docs — markdown / blog content / config / CI workflow files:
-  - Make the change; sanity-check links, frontmatter, and formatting. No app build, browser, or e2e ceremony is required. Run `pnpm typecheck` / `pnpm lint` only if you touched TS/JS.
+  - Make the change; sanity-check links, frontmatter, and formatting. No app build, browser, or e2e ceremony is required. Run `bun run typecheck` / `bun run lint` only if you touched TS/JS.
 - question — a question or code explanation with no file change:
   - Answer concisely and accurately, citing `file:line`. Do not modify code unless the user asks you to.
 
@@ -33,7 +33,7 @@ A browser MCP server is always wired (see "Routing" for which). Browsers and the
 
 - Playwright MCP (server "playwright"): DOM interactions, screenshots, and coordinate-based mouse wheel + drag (needed for D3 zoom/pan). Console via `mcp__playwright__browser_console_messages`, network via `mcp__playwright__browser_network_requests`, JS via `mcp__playwright__browser_evaluate`.
 - Chrome DevTools MCP (server "chrome"): uid-driven. Call `mcp__chrome__take_snapshot` BEFORE click/fill/hover (uids change after navigation — re-snapshot if one fails); screenshot with `take_screenshot`; debug with `list_console_messages` / `list_network_requests` / `evaluate_script`.
-- If the dev server shows as "not started" but your task needs a browser: run `npx -y playwright install --with-deps chromium`, then `pnpm run dev > /tmp/next-dev.log 2>&1 &`, then wait for `curl -sSf http://localhost:3000`.
+- If the dev server shows as "not started" but your task needs a browser: run `bunx playwright install --with-deps chromium`, then `bun run dev > /tmp/next-dev.log 2>&1 &`, then wait for `curl -sSf http://localhost:3000`.
 - The app runs at http://localhost:3000. For docs or external URLs, use WebFetch.
 
 ## Workflow

@@ -4,9 +4,9 @@ You are reviewing code for the InferenceX App — a Next.js frontend dashboard f
 
 The PR's code is checked out and a Next.js dev server has been started (best-effort) at `http://localhost:3000`, backed by the **real read-only database** so charts render real data; its status is in the `dev server:` line at the very top of this prompt. **Don't just read the diff — verify the changed behavior actually works at runtime.** This runtime check is the thing the regular CI can't do; plain correctness is already covered (see step 3).
 
-1. If the `dev server:` line is not `true`, start it yourself and wait: `pnpm run dev > /tmp/next-dev.log 2>&1 &` then poll `curl -sSf http://localhost:3000`. If it still won't start, `tail -n 200 /tmp/next-dev.log`, report that as a 🔴 BLOCKING finding, and fall back to a static review.
+1. If the `dev server:` line is not `true`, start it yourself and wait: `bun run dev > /tmp/next-dev.log 2>&1 &` then poll `curl -sSf http://localhost:3000`. If it still won't start, `tail -n 200 /tmp/next-dev.log`, report that as a 🔴 BLOCKING finding, and fall back to a static review.
 2. Use the **Playwright MCP** (`mcp__playwright__*`) to load the screens and flows this PR changes and confirm they work **as the PR intends**: charts render real data (no "No data available" / "Please change the model…"), the changed interactions/filters/zoom/tooltips behave, and there are no blocking console errors (`mcp__playwright__browser_console_messages`). Save screenshots as evidence. For inference/evaluation changes, also verify the `?unofficialrun=<github-actions-run-id>` overlay path (see AGENTS.md §"Unofficial Run Support").
-3. Do NOT re-run the test suite. The dedicated CI workflows (`tests-unit`, `tests-e2e`, `lint`) already run `typecheck` / `lint` / `test:unit` and the fixtures-based Cypress e2e on this PR. Check their status with the GitHub CI tools (`mcp__github_ci__*`) or `gh pr checks`; if any is failing, pull the failing output and fold it into the review as a 🔴 BLOCKING finding. Do **not** run `pnpm test:e2e` here — Cypress integration needs a fixtures build (`E2E_FIXTURES=1` + `pnpm start`), not this real-data dev server, so it would fail spuriously. Use `Bash` only for targeted investigation of a specific failure.
+3. Do NOT re-run the test suite. The dedicated CI workflows (`tests-unit`, `tests-e2e`, `lint`) already run `typecheck` / `lint` / `test:unit` and the fixtures-based Cypress e2e on this PR. Check their status with the GitHub CI tools (`mcp__github_ci__*`) or `gh pr checks`; if any is failing, pull the failing output and fold it into the review as a 🔴 BLOCKING finding. Do **not** run `bun run test:e2e` here — Cypress integration needs a fixtures build (`E2E_FIXTURES=1` + `bun run start`), not this real-data dev server, so it would fail spuriously. Use `Bash` only for targeted investigation of a specific failure.
 4. Fold what you find — runtime breakage, console errors, charts that don't render, failing CI — into the review alongside the static findings below. Anchor inline comments to the responsible lines in the diff.
 
 For changes with no UI surface (DB/ETL/lib/config), browser verification may not apply — lean on the static review and the CI status from step 3.

@@ -25,33 +25,40 @@ jobs:
       - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
-      - uses: pnpm/action-setup@0ebf47130e4866e96fce0953f49152a61190b271 # v6.0.9
+      - name: Setup Bun
+        id: setup-bun
+        uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2.2.0
+        with:
+          bun-version-file: package.json
       - uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
         with:
           node-version: '24'
-          cache: pnpm
+      - name: Cache Bun packages
+        uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
+        with:
+          path: ~/.bun/install/cache
+          key: ${{ runner.os }}-bun-${{ runner.arch }}-${{ steps.setup-bun.outputs.bun-version }}-${{ hashFiles('bun.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-bun-${{ runner.arch }}-${{ steps.setup-bun.outputs.bun-version }}-
       - name: Install dependencies
-        run: >-
-          pnpm install --frozen-lockfile
-          --filter @semianalysisai/inferencex-db...
-          --filter @semianalysisai/inferencex-app...
+        run: bun install --frozen-lockfile
         env:
           CYPRESS_INSTALL_BINARY: '0'
 
       - name: Apply run overrides
         env:
           DATABASE_WRITE_URL: ${{ secrets.DATABASE_WRITE_URL }}
-        run: pnpm admin:db:apply-overrides --yes
+        run: bun run admin:db:apply-overrides --yes
 
       - name: Verify database
         env:
           DATABASE_WRITE_URL: ${{ secrets.DATABASE_WRITE_URL }}
-        run: pnpm admin:db:verify
+        run: bun run admin:db:verify
 
       - name: Invalidate production cache
         env:
           INVALIDATE_SECRET: ${{ secrets.VERCEL_INVALIDATE_SECRET }}
-        run: pnpm admin:cache:invalidate https://inferencex.semianalysis.com
+        run: bun run admin:cache:invalidate https://inferencex.semianalysis.com
 
       - name: Warm production cache
-        run: pnpm admin:cache:warmup https://inferencex.semianalysis.com
+        run: bun run admin:cache:warmup https://inferencex.semianalysis.com