shikharbhardwaj

4 ms·

1.
Stealing Reasoning Traces from Proprietary LLM APIs(arxiv.org)
6 points by shikharbhardwaj 2mo ago|0 comments
2.
Firstmate: Talk to one agent. Ship with a crew(github.com)
5 points by shikharbhardwaj 2mo ago|0 comments
3.
InboxWhiz: Declutter Your Gmail Inbox(github.com)
2 points by shikharbhardwaj 11mo ago|0 comments
4.
PostgreSQL Is Enough(gist.github.com)
2 points by shikharbhardwaj 11mo ago|0 comments
5.
Software Update Deletes Everything Older Than 10 Days [video](youtube.com)
2 points by shikharbhardwaj 11mo ago|0 comments
6.
Rebuilding Uber's Apache Pinot Query Architecture(uber.com)
3 points by shikharbhardwaj 1y ago|0 comments
7.
Are We Trek Yet?(nicolaiarocci.com)
2 points by shikharbhardwaj 1y ago|1 comments
8.
Preparing for the Long Winter Ahead(ashleyjanssen.com)
3 points by shikharbhardwaj 1y ago|0 comments
9.
Localization in Ruby on Rails: Yes/No, On/Off(allaboutcoding.ghinda.com)
3 points by shikharbhardwaj 1y ago|0 comments
10.
by shikharbhardwaj 1y ago
Hi! Author of the blog post here. I completely agree, getting an objective measure for the developer experience from these various tools is not easy. On one hand, you have a series of benchmarks from LLM providers. While reflecting some deg
11.
Qwen/QwQ-32B released on Hugging Face(huggingface.co)
4 points by shikharbhardwaj 2y ago|0 comments
12.
Supabase is deprecating Fly’s managed Postgres offering on March 14, 2025(github.com)
1 points by shikharbhardwaj 2y ago|0 comments
13.
How I use LLMs by Andrej Karpathy [video](youtube.com)
2 points by shikharbhardwaj 2y ago|0 comments
14.
Building on Quicksand(arxiv.org)
1 points by shikharbhardwaj 2y ago|0 comments
15.
by shikharbhardwaj 2y ago
A few takeaways from this talk: 1. Workflow cost optimization has many inputs and constraints, and good solutions come up gradually when considering the tradeoffs between time, cost and risk. 2. The overall goal should be reducing the aggre
16.
ACM Tech Talk: Tradeoffs in the Software Workflow(youtube.com)
1 points by shikharbhardwaj 2y ago|1 comments
17.
by shikharbhardwaj 2y ago
Git works well, but you don't get realtime/live sync. For my setup, I use the obsidian-livesync [1] plugin with a self-hosted CouchDB instance which has worked well for me. I periodically do commit things to a git repo as a backup
18.
by shikharbhardwaj 2y ago
I run a hybrid setup which has worked well for me: HDDs in the NAS for high-capacity, decent-speed persistent storage with ZFS for redundancy, low-capacity SSDs in the VM/container hosts for speed and reliability.
19.
by shikharbhardwaj 2y ago
It started out as intermittent I/O errors, where the system would boot fine and run for a few hours and then lock up to the point where I could not even SSH into it. I tried replacing the cables and connecting to a different power cabl
20.
by shikharbhardwaj 2y ago
My experience is anecdotal, but as a general consumer using these drives in a home environment, HDDs simply have not been kind to me over the years. On the other hand, I just had my first SSD (Samsung 850 Evo) fail on me yesterday, after ch
21.
by shikharbhardwaj 2y ago
We're already seeing reports of the 12HPWR connector not being able to handle the power the 5090 demands [1]. It seems that for this generation, NVIDIA left little headroom. Might be best to run these undervolted to have some of that h
22.
Nvidia GeForce RTX 5090 Linux GPU Compute Benchmarks(phoronix.com)
21 points by shikharbhardwaj 2y ago|8 comments
23.
State of my Homelab: 2024H1, Part 2(blog.shkhr.dev)
3 points by shikharbhardwaj 2y ago|0 comments
24.
Phoronix Nvidia GeForce RTX 5080 benchmarks(phoronix.com)
2 points by shikharbhardwaj 2y ago|0 comments
25.
by shikharbhardwaj 2y ago
Agreed on CUDA being a big moat; Nvidia has spent a big chunk of the last 2 decades building this ecosystem and is now reaping dividends from it. From what I've read, most of the investments by FAANGs/startups in building speciali
26.
Qwen2.5-VL Is Released(qwenlm.github.io)
5 points by shikharbhardwaj 2y ago|0 comments