Shekhar Gulati
shekhargulati.com · Personal blog · on HN since 2015
Shekhar Gulati, a software engineer and CTO, writes about database design, architecture, programming languages and building tools with LLMs.
What HN saysThe comments give no opinion of the blog; HN links it for notes on Postgres schema design and for his posts on AI tools.
16HN people shared it
2linked it in comments
103stories submitted
858story points
10 months agolast shared
2015people per year2026
Most-shared posts
- 1My notes on Gitlab's Postgres schema design (2022)5 people · 515 points · first shared 2022 · 161 comments on HN
- 2Why an architecture with three Availability Zones?3 people · 6 points · first shared 2022 · discussion on HN
- 3Being Chief Technology Officer: Lessons learned in my first year2 people · 26 points · first shared 2021 · 1 comments on HN
- 4Building a YouTube Video Summarizer with LLM and yt-dlp2 people · 17 points · first shared 2024 · 4 comments on HN
- 5Questions to Ask When Choosing a Programming Language2 people · 7 points · first shared 2021 · discussion on HN
- 6Always Be Reading2 people · 5 points · first shared 2018 · discussion on HN
- 7Why naming stuff is hard?2 people · 4 points · first shared 2022 · discussion on HN
- 8Google AI Overview Has a Timezone Bug2 people · 3 points · first shared 2025 · discussion on HN
- 9My Thoughts on Monorepo1 people · 37 points · first shared 2020 · 18 comments on HN
- 10Why it’s hard for programmers to write a program to flatten a list?1 people · 33 points · first shared 2017 · 120 comments on HN
- 11The 5 minute introduction to Osquery1 people · 9 points · first shared 2019 · discussion on HN
- 12Things I Learnt by Reading Retrofit Source Code1 people · 8 points · first shared 2017 · discussion on HN
- 13On Reading, Creativity, and Self-Reflection1 people · 5 points · first shared 2021 · discussion on HN
- 14Questions to ask when making Build versus Buy decision – Shekhar Gulati1 people · 5 points · first shared 2021 · discussion on HN
- 15Podman: The Missing Tutorial1 people · 5 points · first shared 2021 · discussion on HN
What HN people wrote when they linked it
shekhargulati · 2025-03-07
Mistral OCR made multiple mistakes in extracting this [1] document. It is a two-page-long PDF in Arabic from the Saudi Central Bank. The following errors were observed:
- Referenced Vision 2030 as Vision 2.0. - Failed to extract the table; instead, it hallucinated and extracted the text in a different format. - Failed to extract the number and date of the circular.
I tested the same document with ChatGPT, Claude, Grok, and Gemini. Only Claude 3.7 extracted the complete document, while all others failed badly. You can read my analysis here [2].
1. https://rulebook.sama.gov.sa/sites/default/files/en_net_file_store/SAM… 2. https://shekhargulati.com/2025/03/05/claude-3-7-sonnet-is-good-at-pdf-…
rrdharan · 2020-11-26
They said they were the fastest.
They shipped with syncs turned off by default.
EDIT: actually it was journaling of writeconcerns. Later they backtracked and enabled it by default, but by then their reputation was cemented (or bifurcated into fast, or terrible, I suppose).
https://shekhargulati.com/2011/12/08/how-mongodb-different-write-conce…