Hacker News: swaminarayan

New comment by swaminarayan in "Show HN: Find the best local LLM for your hardware, ranked by benchmarks"

swaminarayan — Fri, 15 May 2026 14:34:05 +0000

How it select model? using AI?

New comment by swaminarayan in "Show HN: Query months of HN discussions and comments as a knowledge graph"

swaminarayan — Tue, 12 May 2026 14:36:20 +0000

I am getting used to with this lightrag-snkv, Now I dont need to check hacker news site again, it automatically ingest top ranked posts and I can query and check answers using natural query language.

New comment by swaminarayan in "Show HN: Query months of HN discussions and comments as a knowledge graph"

swaminarayan — Mon, 11 May 2026 01:44:57 +0000

I am pasting link here: https://github.com/user-attachments/assets/23a95319-ee66-49b...

New comment by swaminarayan in "Show HN: Query months of HN discussions and comments as a knowledge graph"

swaminarayan — Mon, 11 May 2026 01:42:59 +0000

if you are opening video in mobile github app then it won’t open.Check in browser.

New comment by swaminarayan in "Show HN: Query years of HN discussions and comments as a knowledge graph"

swaminarayan — Sun, 10 May 2026 09:55:24 +0000

Right now Show HN and Ask HN tags based blogs and its comments are ingested in database.

Show HN: Query months of HN discussions and comments as a knowledge graph

swaminarayan — Sun, 10 May 2026 08:46:59 +0000

I built lightrag-snkv, Basically it uses lightRAG https://github.com/HKUDS/LightRAG ,this requires various storage databases like key value store, graph database, vector database, I built single embedded file based database which covers all these requirements: https://github.com/hash-anu/snkv. And developed ingestor which ingest data from Algolia's HN search API. Here auto ingestion timings can be configured either using server.py or it can be configured by ingester code file or can be configured in env file.

This is very early stage and you might hit issues.

Demo video and quick-start in the README. Let me know if you run into issues.

Comments URL: https://news.ycombinator.com/item?id=48082150

Points: 4

# Comments: 5

New comment by swaminarayan in "Show HN: Ithihāsas – a character explorer for Hindu epics, built in a few hours"

swaminarayan — Tue, 14 Apr 2026 02:39:05 +0000

this is not mythology. this is ithihasas meaning thus it happened

New comment by swaminarayan in "Ask HN: Best LLM model for a RAG-based Android app across all smartphones?"

swaminarayan — Fri, 03 Apr 2026 10:54:47 +0000

Found my answer: https://github.com/google-ai-edge/gallery?tab=readme-ov-file

Ask HN: Best LLM model for a RAG-based Android app across all smartphones?

swaminarayan — Fri, 03 Apr 2026 07:22:31 +0000

I am developing RAG based android app using llama.cpp. For offline processing I am using Qwen 1.5 2.5B model using Q4 quantization, I am offloading computation to GPU if present, However for low end android phones my android app is either crashing due to OOM or If not crashing and no GPU available in that case it takes lots of time to generate text. I also tried with SmolLM 135M model for low end devices but it struggle to follow instruction very well.

In this case I am considering openAI API for low end android phones. And for vector storage I am using in house developed https://github.com/hash-anu/snkv,

I am not sure how other people are running LLM model on low end android devices, I would appreciate any insights or best practices.

Comments URL: https://news.ycombinator.com/item?id=47624065

Points: 1

# Comments: 1

New comment by swaminarayan in "Show HN: Dull – Instagram Without Reels, YouTube Without Shorts (iOS)"

swaminarayan — Thu, 02 Apr 2026 08:11:01 +0000

i like this idea, especially for the parent who dont want their kids to watch reels/shorts of the instagram and youtube apps.

New comment by swaminarayan in "Show HN: Postgres extension for BM25 relevance-ranked full-text search"

swaminarayan — Thu, 02 Apr 2026 03:19:58 +0000

lol

New comment by swaminarayan in "Show HN: Postgres extension for BM25 relevance-ranked full-text search"

swaminarayan — Wed, 01 Apr 2026 17:34:45 +0000

What architectural decisions enabled pg_textsearch to outperform Tantivy/ParadeDB by 4.7×, especially given Tantivy’s reputation for speed?

New comment by swaminarayan in "DoltLite: Free, open source drop-in replacement for SQLite with version control"

swaminarayan — Fri, 27 Mar 2026 12:25:18 +0000

Interesting approach — especially around modifying SQLite’s internals.

In SNKV, I’m taking almost the opposite direction: keeping B-tree / storage-layer ideas and removing the SQL layer entirely, to expose a minimal key-value interface.

Curious how you think about the tradeoff between simplifying the query layer vs the storage layer.

https://github.com/hash-anu/snkv

New comment by swaminarayan in "In-process reimplementation of PostgreSQL backed by SQLite-compatible storage"

swaminarayan — Fri, 27 Mar 2026 12:12:13 +0000

Really interesting design — especially using SQLite-compatible storage with a PostgreSQL interface.

I'm working on SNKV, which explores a different approach: skipping the SQL layer entirely and focusing on a simple key-value interface.

It works closer to the storage layer (b-tree / WAL / pager concepts), aiming for a lightweight and minimal design, while supporting multi-process access.

Curious how you think about the tradeoff between SQL flexibility and a simpler KV interface.

https://github.com/hash-anu/snkv

New comment by swaminarayan in "Show HN: Turbolite – a SQLite VFS serving sub-250ms cold JOIN queries from S3"

swaminarayan — Fri, 27 Mar 2026 03:54:20 +0000

Very cool approach. I’m working on a related KV storage project (snkv - https://github.com/hash-anu/snkv)

How do you handle manifest consistency and updates—append-only or rewrite-heavy?

New comment by swaminarayan in "Show HN: I took back Video.js after 16 years and we rewrote it to be 88% smaller"

swaminarayan — Wed, 25 Mar 2026 09:48:21 +0000

What were the biggest architectural changes in the rewrite, and what tradeoffs did you make compared to the old Video.js design?

New comment by swaminarayan in "Show HN: DuckDB community extension for prefiltered HNSW using ACORN-1"

swaminarayan — Wed, 25 Mar 2026 09:43:37 +0000

Does your method work better than standard ANN when filters are very strict—and how does it affect speed vs accuracy?

New comment by swaminarayan in "Show HN: Termcraft – Terminal-first 2D sandbox survival in Rust"

swaminarayan — Sun, 22 Mar 2026 06:44:59 +0000

How do you build a procedurally generated survival game in a terminal without a graphics engine?

New comment by swaminarayan in "Show HN: Three new Kitten TTS models – smallest less than 25MB"

swaminarayan — Fri, 20 Mar 2026 02:38:47 +0000

How did you make a very small AI model (14M) sound more natural and expressive than even bigger models?

New comment by swaminarayan in "Show HN: Pgit – A Git-like CLI backed by PostgreSQL"

swaminarayan — Thu, 19 Mar 2026 02:57:38 +0000

if you want to use key value store using sqlite then you can try : https://github.com/hash-anu/snkv

in which i am directly accessing b-tree layer and not consuming query layer.

for kv workloads it is much faster compare to sql.

and yeah you will get same benefits of sqlite storage engine.