---
title: "Jeet Dhandha (@jeetdhandha) | Hashnode"
description: "Jeet Dhandha (@jeetdhandha) writes on Hashnode. Read their latest articles and follow their work."
url: "https://hashnode.com/@jeetdhandha"
---

@jeetdhandha

# Jeet Dhandha

@jeetdhandha · Joined July 2022

### [An ML search loop that can't overfit its own eval — the gate is in code, not a prompt](https://jd-dev.hashnode.dev/an-ml-search-loop-that-can-t-overfit-its-own-eval-the-gate-is-in-code-not-a-prompt?utm_source=hashnode&utm_medium=feed)

I kept seeing agent and eval demos where the honesty — held-out discipline, "no metric gaming" — lives in a prompt, or in a paper's methodology section. So I tried to build the opposite: a search loop

3 min read · [jd-dev.hashnode.dev](https://jd-dev.hashnode.dev/an-ml-search-loop-that-can-t-overfit-its-own-eval-the-gate-is-in-code-not-a-prompt?utm_source=hashnode&utm_medium=feed)

[1](https://hashnode.com/posts/an-ml-search-loop-that-can-t-overfit-its-own-eval-the-gate-is-in-code-not-a-prompt/6a64a424431a612521b53833) [Jul 25](https://hashnode.com/posts/an-ml-search-loop-that-can-t-overfit-its-own-eval-the-gate-is-in-code-not-a-prompt/6a64a424431a612521b53833)

### [qs-fast: a drop-in faster qs (2.3× parse, 3.8× stringify, identical output)](https://jd-dev.hashnode.dev/qs-fast-a-drop-in-faster-qs-2-3-parse-3-8-stringify-identical-output?utm_source=hashnode&utm_medium=feed)

A drop-in replacement for qs that fast-paths the common case — 2.3× faster parse, 3.8× faster stringify on V8 — with byte-for-byte identical output and a safe fallback.

2 min read · [jd-dev.hashnode.dev](https://jd-dev.hashnode.dev/qs-fast-a-drop-in-faster-qs-2-3-parse-3-8-stringify-identical-output?utm_source=hashnode&utm_medium=feed)

[Jul 17](https://hashnode.com/posts/qs-fast-a-drop-in-faster-qs-2-3-parse-3-8-stringify-identical-output/6a5a3226f2b38c87c10ce491)

### [diff-fast: a drop-in faster jsdiff (2.6–4.8×, identical output)](https://jd-dev.hashnode.dev/diff-fast-a-drop-in-faster-jsdiff-2-6-4-8-identical-output?utm_source=hashnode&utm_medium=feed)

A drop-in replacement for jsdiff that makes diffChars/diffLines 2.6–4.8× faster on V8 with byte-for-byte identical output.

3 min read · [jd-dev.hashnode.dev](https://jd-dev.hashnode.dev/diff-fast-a-drop-in-faster-jsdiff-2-6-4-8-identical-output?utm_source=hashnode&utm_medium=feed)

[Jul 17](https://hashnode.com/posts/diff-fast-a-drop-in-faster-jsdiff-2-6-4-8-identical-output/6a5a2fea091a449c2ff7d6e2)

### [Stop re-parsing your font 100,000 times: faster bulk PDF generation with pdf-lib-bulk](https://jd-dev.hashnode.dev/stop-re-parsing-your-font-100-000-times-faster-bulk-pdf-generation-with-pdf-lib-bulk?utm_source=hashnode&utm_medium=feed)

pdf-lib re-parses your font on every embedFont() call. pdf-lib-bulk caches the parse once and generates PDFs up to 4x faster on variable documents and 31x on identical ones. Zero deps, MIT.

6 min read · [jd-dev.hashnode.dev](https://jd-dev.hashnode.dev/stop-re-parsing-your-font-100-000-times-faster-bulk-pdf-generation-with-pdf-lib-bulk?utm_source=hashnode&utm_medium=feed)

[Jul 16](https://hashnode.com/posts/stop-re-parsing-your-font-100-000-times-faster-bulk-pdf-generation-with-pdf-lib-bulk/6a5951575f9d63c9cca1026f)
