fieldkit 0.1.0__tar.gz

fieldkit-0.1.0/.gitignore

@@ -0,0 +1,48 @@
+# build output
+dist/
+.astro/
+
+# dependencies
+node_modules/
+
+# logs
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+pnpm-debug.log*
+
+# environment variables
+.env
+.env.production
+
+# macOS-specific files
+.DS_Store
+
+# jetbrains setting folder
+.idea/
+
+# local-only working material (not for the public blog)
+ideas/
+HANDOFF.md
+.claude/
+
+# transient vibe-test artifacts (Playwright screenshots written to repo root)
+.playwright-mcp/
+/*.png
+
+# bench-run outputs land inside articles/<slug>/evidence/runs/ — bulky, not for git
+articles/*/evidence/runs/
+
+__pycache__/
+
+# fieldkit Python build / venv detritus (keep alongside the package, not committed)
+fieldkit/build/
+fieldkit/dist/
+fieldkit/*.egg-info/
+fieldkit/.pytest_cache/
+fieldkit/.mypy_cache/
+fieldkit/.ruff_cache/
+fieldkit/.venv/
+fieldkit/src/fieldkit.egg-info/
+*.pyc
+

fieldkit-0.1.0/CHANGELOG.md

@@ -0,0 +1,41 @@
+# Changelog
+
+All notable changes to `fieldkit` are documented in this file.
+
+The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and the project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). While the package is on `0.x`, minor versions may include breaking changes. `1.0` will mark API stability.
+
+## [Unreleased]
+
+## [0.1.0] — 2026-05-02
+
+First public release. Four library modules + a CLI + a docs site section, lifted from 25+ articles in [ai-field-notes](https://ainative.business/field-notes/).
+
+### Added
+
+- **`fieldkit.capabilities`** — typed Python facade over `spark-capabilities.json`. `Capabilities.load()` cached singleton (with `.hardware`, `.memory_budget_rules_of_thumb`, `.stack`, `.in_envelope_signals`, `.out_of_envelope_signals`, `.stage_routing_hints`, `.series_routing_hints`), plus canonical math helpers `kv_cache_bytes()`, `weight_bytes()`, `practical_inference_envelope()`. Numbers pinned to `kv-cache-arithmetic-at-inference` and `gpu-sizing-math-for-fine-tuning`. ([#capabilities])
+- **`fieldkit.nim`** — OpenAI-compatible `NIMClient` over `httpx` with `tenacity`-backed retries on 429 / 503 / `ConnectError` / timeouts. `NIMClient.chat()` runs a pre-flight context check and raises `NIMContextOverflowError` with the estimated token count *before any network call*, so the opaque NIM 400 from a >8192-token request never surfaces. Helpers: `chunk_text()` (paragraph→sentence→word splitting under a `max_tokens` budget), `estimate_tokens()` (1 tok ≈ 4 chars), `wait_for_warm()` (polls `/v1/models` for the ~90s NIM cold start). Constants: `NIM_CONTEXT_WINDOW = 8192`, `DEFAULT_CHUNK_TOKENS = 1024`. Errors: `NIMError` → `NIMHTTPError`, `NIMTimeoutError`, `NIMContextOverflowError`. ([#nim])
+- **`fieldkit.rag`** — composable ingest → retrieve → rerank → fuse pipeline backed by pgvector + a NIM embedder + the strict-context grounded prompt from `naive-rag-on-spark`. `Pipeline.ingest()` chunks via `fieldkit.nim.chunk_text` and upserts in batches of 32; `Pipeline.retrieve()` does pgvector cosine top-K; `Pipeline.rerank()` is a pass-through when `rerank_url=None`; `Pipeline.fuse()` builds the strict-context messages list and calls the generator; `Pipeline.ask()` chains all three. Embed and rerank inherit `NIMClient.chat`'s retry policy so co-resident memory pressure doesn't fail the pipeline. ([#rag])
+- **`fieldkit.eval`** — `Bench` (latency aggregation with the same `{summary, calls}` JSON shape as the article evidence files), `Judge` (LLM-as-judge with built-in `correctness` / `faithfulness` / `relevance` rubrics + a static `Judge.parse()` JSON-then-regex extractor), `Trajectory` (agent-loop JSONL analyzer with `knob_coverage / repeat_rate / mode_dominance / cumulative_best`), `is_refusal()` (regex catalog unioned across the project's articles), `summarize_metric()`. ([#eval])
+- **`fieldkit.cli`** — Typer wrapper exposing `fieldkit version`, `fieldkit envelope <size>`, `fieldkit feasibility <model_id> [--ctx --batch --dtype]`, `fieldkit bench rag`. On `$PATH` after `pip install`. ([#cli])
+- **Astro docs site** — `/fieldkit/` landing page with install + quickstart + module grid, and `/fieldkit/api/<module>/` reference pages backed by a new `fieldkit_docs` content collection. Articles can opt-in via `fieldkit_modules:` frontmatter to display a "USES fieldkit.X" chip on cards and appear under each module's "Articles that use fieldkit.<module>" footer. 11 articles opted in for the v0.1.0 launch.
+- **Samples**: `samples/feasibility-math.py` (capabilities reproduction of the kv-cache article's table), `samples/hello-nim.py` (Python equivalent of the curl one-liner), `samples/naive-rag.py` (end-to-end RAG in <30 lines), `samples/bench-rag.py` (offline `Bench` + `Judge.parse()` walkthrough).
+- `scripts/sync_capabilities.py` keeps the package-bundled `spark-capabilities.json` in sync with the source-of-truth at `scripts/lib/spark-capabilities.json` (pre-commit-enforced).
+- `pytest --spark` flag (via `tests/conftest.py`) gates integration tests that need a live NIM / pgvector on the DGX Spark; default runs skip them.
+
+### Changed
+
+- `frontier-scout` skill (`refresh` and `eval` modes, plus `references/feasibility-prompt.md` and `references/classifier-prompt.md`) now teaches the typed `from fieldkit.capabilities import …` API as the preferred grounding path; raw JSON read is the documented fallback.
+
+### Verified on Spark
+
+Phases 3, 4, 5 were live-verified end-to-end against the chat NIM (Llama 3.1 8B, port 8000), the embed NIM (Nemotron Embed 1B v2, port 8001), and pgvector (port 5432) before being committed. Phase 5 in particular rewrote `articles/naive-rag-on-spark/evidence/benchmark.py` against `fieldkit.eval.Bench` + `fieldkit.rag.Pipeline.fuse` and reproduced the original article's behavioral fingerprint (5 of 6 refusals incl. the canonical Google-IPO false refusal, plus the Ian Thorpe grounded answer).
+
+### Test suite
+
+**157 passing, 2 skipped** without `--spark` (151 passing with `--spark` against warm NIMs + pgvector). Reproduce: `pip install fieldkit/[dev]` then `pytest`; for the live tests, `pytest --spark`.
+
+[#capabilities]: https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit/src/fieldkit/capabilities
+[#nim]: https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit/src/fieldkit/nim
+[#rag]: https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit/src/fieldkit/rag
+[#eval]: https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit/src/fieldkit/eval
+[#cli]: https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit/src/fieldkit/cli

fieldkit-0.1.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

fieldkit-0.1.0/PKG-INFO

@@ -0,0 +1,92 @@
+Metadata-Version: 2.4
+Name: fieldkit
+Version: 0.1.0
+Summary: Verified-on-Spark patterns lifted from the ai-field-notes blog into one importable Python package.
+Project-URL: Homepage, https://ainative.business/fieldkit/
+Project-URL: Source, https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit
+Project-URL: Changelog, https://github.com/manavsehgal/ai-field-notes/blob/main/fieldkit/CHANGELOG.md
+Project-URL: Issues, https://github.com/manavsehgal/ai-field-notes/issues
+Author-email: Manav Sehgal <sehgal.manav@gmail.com>
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: agents,dgx-spark,llm,nim,nvidia,rag
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.11
+Requires-Dist: httpx>=0.27
+Requires-Dist: numpy>=1.26
+Requires-Dist: psycopg[binary]>=3.2
+Requires-Dist: tenacity>=8.5
+Requires-Dist: typer>=0.12
+Provides-Extra: dev
+Requires-Dist: build>=1.2; extra == 'dev'
+Requires-Dist: mypy>=1.11; extra == 'dev'
+Requires-Dist: pre-commit>=3.7; extra == 'dev'
+Requires-Dist: pytest-asyncio>=0.23; extra == 'dev'
+Requires-Dist: pytest>=8.2; extra == 'dev'
+Requires-Dist: respx>=0.21; extra == 'dev'
+Requires-Dist: ruff>=0.6; extra == 'dev'
+Requires-Dist: twine>=5.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# fieldkit
+
+> Verified-on-Spark patterns lifted from the [ai-field-notes](https://ainative.business/field-notes/) blog into one importable Python package.
+
+Every essay in `ai-field-notes` ends with `evidence/` — a folder of working code that produced the article's numbers. After 25+ articles the same patterns kept reappearing: the same NIM client wrapper, the same chunk-embed-store dance, the same bench harness. `fieldkit` is what those `evidence/` folders look like once the boilerplate is lifted into a real package.
+
+The blog stays the long-form rationale. `fieldkit` is the `pip install`-able surface so you can reproduce — and extend — the work without re-pasting 80 lines of NIM-client setup per article.
+
+## Install
+
+```bash
+pip install fieldkit
+```
+
+For the bleeding edge between releases, install from the git tag instead:
+
+```bash
+pip install "git+https://github.com/manavsehgal/ai-field-notes.git@fieldkit/v0.1.0#subdirectory=fieldkit"
+```
+
+## Quickstart
+
+```python
+from fieldkit.nim import NIMClient
+
+client = NIMClient(base_url="http://localhost:8000/v1", model="meta/llama-3.1-8b-instruct")
+print(client.chat([{"role": "user", "content": "Hello, Spark."}]))
+```
+
+## What's in v0.1.0
+
+| Module | Purpose | Source articles |
+|---|---|---|
+| `fieldkit.capabilities` | Typed Python facade over `spark-capabilities.json` — KV cache math, weight bytes, inference envelope. | `kv-cache-arithmetic-at-inference`, `gpu-sizing-math-for-fine-tuning` |
+| `fieldkit.nim` | OpenAI-compatible NIM client wrapper with retry, chunking, and the 8192-token context guard. | `nim-first-inference-dgx-spark` and friends |
+| `fieldkit.rag` | `Pipeline(embed_url, rerank_url, pgvector_dsn, generator)` — ingest → retrieve → rerank → fuse. | `naive-rag-on-spark` and friends |
+| `fieldkit.eval` | `Bench`, `Judge`, `Trajectory` — the recurring eval harness shapes. | every article with a `bench.py` or `benchmark.py` |
+| `fieldkit.cli` | `fieldkit bench rag`, `fieldkit feasibility <id>`, `fieldkit envelope <size>`. | discoverability |
+
+Modules deferred to `v0.2`: `retriever`, `ft`, `guardrails`, `agents`. To `v0.3`: `train`, `observe`.
+
+## Hardware
+
+`v0.1` is **Spark-only**. Every code path is verified on a DGX Spark (GB10, 128 GB unified memory, NIM 8B + embed NIM + pgvector co-resident). Portability to other CUDA 12.x boxes lands in `v0.2+` when there's demand.
+
+## License
+
+Apache-2.0. See [`LICENSE`](https://github.com/manavsehgal/ai-field-notes/blob/main/fieldkit/LICENSE).
+
+## Links
+
+- **Blog:** https://ainative.business/field-notes/
+- **Docs:** https://ainative.business/fieldkit/
+- **Source:** https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit
+- **Changelog:** [`CHANGELOG.md`](https://github.com/manavsehgal/ai-field-notes/blob/main/fieldkit/CHANGELOG.md)

fieldkit-0.1.0/README.md

@@ -0,0 +1,55 @@
+# fieldkit
+
+> Verified-on-Spark patterns lifted from the [ai-field-notes](https://ainative.business/field-notes/) blog into one importable Python package.
+
+Every essay in `ai-field-notes` ends with `evidence/` — a folder of working code that produced the article's numbers. After 25+ articles the same patterns kept reappearing: the same NIM client wrapper, the same chunk-embed-store dance, the same bench harness. `fieldkit` is what those `evidence/` folders look like once the boilerplate is lifted into a real package.
+
+The blog stays the long-form rationale. `fieldkit` is the `pip install`-able surface so you can reproduce — and extend — the work without re-pasting 80 lines of NIM-client setup per article.
+
+## Install
+
+```bash
+pip install fieldkit
+```
+
+For the bleeding edge between releases, install from the git tag instead:
+
+```bash
+pip install "git+https://github.com/manavsehgal/ai-field-notes.git@fieldkit/v0.1.0#subdirectory=fieldkit"
+```
+
+## Quickstart
+
+```python
+from fieldkit.nim import NIMClient
+
+client = NIMClient(base_url="http://localhost:8000/v1", model="meta/llama-3.1-8b-instruct")
+print(client.chat([{"role": "user", "content": "Hello, Spark."}]))
+```
+
+## What's in v0.1.0
+
+| Module | Purpose | Source articles |
+|---|---|---|
+| `fieldkit.capabilities` | Typed Python facade over `spark-capabilities.json` — KV cache math, weight bytes, inference envelope. | `kv-cache-arithmetic-at-inference`, `gpu-sizing-math-for-fine-tuning` |
+| `fieldkit.nim` | OpenAI-compatible NIM client wrapper with retry, chunking, and the 8192-token context guard. | `nim-first-inference-dgx-spark` and friends |
+| `fieldkit.rag` | `Pipeline(embed_url, rerank_url, pgvector_dsn, generator)` — ingest → retrieve → rerank → fuse. | `naive-rag-on-spark` and friends |
+| `fieldkit.eval` | `Bench`, `Judge`, `Trajectory` — the recurring eval harness shapes. | every article with a `bench.py` or `benchmark.py` |
+| `fieldkit.cli` | `fieldkit bench rag`, `fieldkit feasibility <id>`, `fieldkit envelope <size>`. | discoverability |
+
+Modules deferred to `v0.2`: `retriever`, `ft`, `guardrails`, `agents`. To `v0.3`: `train`, `observe`.
+
+## Hardware
+
+`v0.1` is **Spark-only**. Every code path is verified on a DGX Spark (GB10, 128 GB unified memory, NIM 8B + embed NIM + pgvector co-resident). Portability to other CUDA 12.x boxes lands in `v0.2+` when there's demand.
+
+## License
+
+Apache-2.0. See [`LICENSE`](https://github.com/manavsehgal/ai-field-notes/blob/main/fieldkit/LICENSE).
+
+## Links
+
+- **Blog:** https://ainative.business/field-notes/
+- **Docs:** https://ainative.business/fieldkit/
+- **Source:** https://github.com/manavsehgal/ai-field-notes/tree/main/fieldkit
+- **Changelog:** [`CHANGELOG.md`](https://github.com/manavsehgal/ai-field-notes/blob/main/fieldkit/CHANGELOG.md)

fieldkit-0.1.0/docs/api/capabilities.md

@@ -0,0 +1,98 @@
+---
+module: capabilities
+title: fieldkit.capabilities
+summary: Typed Python facade over the project's Spark capabilities map. Canonical KV-cache and weight arithmetic.
+order: 1
+---
+
+## What it is
+
+A read-only typed view of `spark-capabilities.json` — the project's grounding floor for hardware envelope claims (KV-cache math, weight memory, in/out-envelope signals, NIM/NeMo/TRT-LLM stack notes). The same JSON the `frontier-scout` skill uses to decide whether a paper fits on the Spark.
+
+The package keeps its own copy of the JSON in sync with the source-of-truth at `scripts/lib/spark-capabilities.json` via a pre-commit drift check.
+
+## Public API
+
+```python
+from fieldkit.capabilities import (
+    Capabilities,
+    kv_cache_bytes,
+    weight_bytes,
+    practical_inference_envelope,
+    DTYPE_BYTES,
+    UnknownDtype,
+    UnknownEnvelope,
+)
+```
+
+### `Capabilities.load(refresh=False) -> Capabilities`
+
+Cached singleton typed view. Pass `refresh=True` to force a re-read from disk.
+
+```python
+caps = Capabilities.load()
+caps.hardware.unified_memory_gb        # 128
+caps.hardware.compute_arch             # "GB10 Grace Blackwell"
+caps.memory_budget_rules_of_thumb.practical_inference_envelope
+caps.stack["nim"].verified_in_articles
+caps.in_envelope_signals               # tuple[str, ...]
+caps.out_of_envelope_signals
+caps.stage_routing_hints               # {"inference": "...", ...}
+caps.raw                               # full JSON dict for ad-hoc inspection
+```
+
+### `kv_cache_bytes(*, hidden, n_layers, ctx, batch, dtype) -> int`
+
+Canonical KV-cache equation from `kv-cache-arithmetic-at-inference`:
+
+```
+KV bytes = 2 × n_layers × kv_hidden × ctx × batch × bytes_per_dtype
+```
+
+`hidden` here is the **KV hidden size** (`n_kv_heads × head_dim`), not the model's full hidden dim — important for GQA models like Llama 3.1 70B (8 KV heads × 128 head_dim = 1024).
+
+```python
+kv_cache_bytes(hidden=1024, n_layers=80, ctx=16384, batch=32, dtype="fp16")
+# 171_798_691_840  (≈ 171.8 GB)
+```
+
+### `weight_bytes(*, params_b, dtype) -> int`
+
+Weight memory in bytes for `params_b` billion parameters at `dtype`.
+
+```python
+weight_bytes(params_b=70, dtype="bf16")   # 140_000_000_000  (140 GB)
+weight_bytes(params_b=100, dtype="fp8")   # 100_000_000_000  (100 GB)
+weight_bytes(params_b=100, dtype="nf4")   #  50_000_000_000  ( 50 GB)
+```
+
+### `practical_inference_envelope(model_size: str) -> str`
+
+Look up the rule-of-thumb envelope string for a model size.
+
+```python
+practical_inference_envelope("8B params bf16")
+# "fits with room — ~16 GB weights + KV; 24.8 tok/s measured on NIM"
+
+practical_inference_envelope("70B params fp8")
+# "~70 GB weights; leaves ~50 GB for KV + activations + system; tight but possible"
+```
+
+Raises `UnknownEnvelope` if no rule matches.
+
+### `DTYPE_BYTES`
+
+Bytes-per-parameter table:
+
+| dtype | bytes |
+|---|---|
+| `fp32` | 4 |
+| `bf16` / `fp16` | 2 |
+| `fp8` / `int8` | 1 |
+| `int4` / `nf4` | 0.5 |
+
+Unknown dtype → `UnknownDtype`.
+
+## Sample
+
+[`samples/feasibility-math.py`](https://github.com/manavsehgal/ai-field-notes/blob/main/fieldkit/samples/feasibility-math.py) reproduces the kv-cache article's serving table, the 100B Nemotron weight table, and the envelope lookup, all via the public API.

fieldkit-0.1.0/docs/api/cli.md

@@ -0,0 +1,74 @@
+---
+module: cli
+title: fieldkit (CLI)
+summary: A thin Typer wrapper over the four modules. Quick checks and smoke benchmarks without writing Python.
+order: 5
+---
+
+## What it is
+
+A thin command-line entry point exposed at `fieldkit` after `pip install`. Every subcommand is a ~20-line wrapper over the existing module APIs — for real workloads, import `fieldkit.{capabilities,nim,rag,eval}` directly instead.
+
+## Commands
+
+### `fieldkit version`
+
+Print the installed package version.
+
+```bash
+$ fieldkit version
+0.1.0.dev0
+```
+
+### `fieldkit envelope <size>`
+
+Look up the practical inference envelope rule for a model size from `spark-capabilities.json`.
+
+```bash
+$ fieldkit envelope "70B params fp8"
+~70 GB weights; leaves ~50 GB for KV + activations + system; tight but possible
+
+$ fieldkit envelope "8B params bf16"
+fits with room — ~16 GB weights + KV; 24.8 tok/s measured on NIM
+```
+
+Unknown size → exit code 2 with the list of known keys.
+
+### `fieldkit feasibility <model_id>`
+
+Quick weights + KV-cache feasibility view for a known shape. Built-in catalog: `llama-3.1-8b`, `llama-3.1-70b`, `100b-bf16`.
+
+```bash
+$ fieldkit feasibility llama-3.1-70b --ctx 4096 --batch 32 --dtype fp8
+model:           llama-3.1-70b
+hardware:        NVIDIA DGX Spark (GB10 Grace Blackwell) — 128 GB unified
+weights (fp8):       70.0 GB
+KV cache (fp8):      21.5 GB  (ctx=4096, batch=32)
+weights + KV:       91.5 GB
+envelope rule:   ~70 GB weights; leaves ~50 GB for KV + activations + system; tight but possible
+```
+
+Flags:
+
+| Flag | Default | Notes |
+|---|---|---|
+| `--ctx` | 4096 | Context length in tokens |
+| `--batch` | 1 | Concurrency / batch size |
+| `--dtype` | fp16 | `fp32`, `bf16`, `fp16`, `fp8`, `int8`, `int4`, `nf4` |
+
+### `fieldkit bench rag`
+
+Smoke-bench `Pipeline.ask` against a 3-doc in-memory corpus. Requires the chat NIM, embed NIM, and pgvector to be reachable.
+
+```bash
+$ fieldkit bench rag --table fieldkit_cli_bench_rag --out /tmp/bench.json
+waiting for embed NIM at http://localhost:8001/v1 ...
+waiting for chat NIM at http://localhost:8000/v1 ...
+ingested 3 chunks into fieldkit_cli_bench_rag
+
+| call | latency_ms | success |
+|---|---|---|
+| ... |
+```
+
+Env vars: `EMBED_BASE_URL`, `NIM_BASE_URL`, `NIM_MODEL`, `PGVECTOR_DSN`. Each has a matching `--flag` form.