Hacker News: Gusarich

OpenAI Symphony

Gusarich — Wed, 04 Mar 2026 18:50:39 +0000

Article URL: https://github.com/openai/symphony

Comments URL: https://news.ycombinator.com/item?id=47252045

Points: 25

# Comments: 6

New comment by Gusarich in "Claude Opus 4.6"

Gusarich — Thu, 05 Feb 2026 17:41:44 +0000

not out yet

Things got too easy with AI

Gusarich — Tue, 27 Jan 2026 08:02:06 +0000

Article URL: https://gusarich.com/blog/things-got-too-easy

Comments URL: https://news.ycombinator.com/item?id=46776836

Points: 1

# Comments: 0

TON Vanity: 286,000x faster vanity addresses

Gusarich — Fri, 02 Jan 2026 20:20:34 +0000

Article URL: https://gusarich.com/blog/ton-vanity/

Comments URL: https://news.ycombinator.com/item?id=46468935

Points: 2

# Comments: 0

AI in 2026

Gusarich — Wed, 31 Dec 2025 15:25:04 +0000

Article URL: https://gusarich.com/blog/ai-in-2026/

Comments URL: https://news.ycombinator.com/item?id=46444927

Points: 2

# Comments: 0

What LLM to use today?

Gusarich — Fri, 28 Nov 2025 07:11:41 +0000

Article URL: https://gusarich.com/blog/what-llm-to-use-today/

Comments URL: https://news.ycombinator.com/item?id=46076281

Points: 2

# Comments: 0

There is nothing out-of-distribution

Gusarich — Thu, 20 Nov 2025 13:18:20 +0000

Article URL: https://gusarich.com/blog/there-is-nothing-out-of-distribution/

Comments URL: https://news.ycombinator.com/item?id=45992253

Points: 2

# Comments: 0

There is no singularity

Gusarich — Sun, 12 Oct 2025 22:21:46 +0000

Article URL: https://gusarich.com/blog/there-is-no-singularity/

Comments URL: https://news.ycombinator.com/item?id=45562554

Points: 2

# Comments: 1

Show HN: Pitaya – Orchestrate AI coding agents like Claude Code

Gusarich — Sun, 31 Aug 2025 16:33:21 +0000

Pitaya is a local, open-source orchestrator for AI coding agents (Claude Code, Codex CLI). It runs many agents in parallel, isolates each in Docker with its own git branch, supports pluggable Python strategies, and persists state so runs are resumable. Quickstart + short demo are in the README.

Comments URL: https://news.ycombinator.com/item?id=45084477

Points: 5

# Comments: 0

New comment by Gusarich in "A "Livestream" Dashboard for Hacker News – Newest Story and Live Comments"

Gusarich — Mon, 18 Aug 2025 06:05:40 +0000

cool but i'd rather have it as a browser extension for hn and not as a separate website

New comment by Gusarich in "Is AI Hitting a Wall?"

Gusarich — Sun, 17 Aug 2025 05:26:32 +0000

it's not hitting a wall anytime soon. just look at gpt-5.

New comment by Gusarich in "Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens"

Gusarich — Tue, 12 Aug 2025 05:59:59 +0000

The article already seems outdated on the first day. The key points about SFT are irrelevant in the era of RL.

Avatarl: Training language models from scratch with pure reinforcement learning

Gusarich — Sat, 09 Aug 2025 14:51:47 +0000

Article URL: https://tokenbender.com/post.html?id=avatarl

Comments URL: https://news.ycombinator.com/item?id=44846962

Points: 9

# Comments: 0

Accessing GPT-5 in Perplexity

Gusarich — Sat, 02 Aug 2025 14:42:11 +0000

Article URL: https://twitter.com/konovenski/status/1951635206592495921

Comments URL: https://news.ycombinator.com/item?id=44768001

Points: 1

# Comments: 0

New comment by Gusarich in "Ask HN: Claude Code vs. ChatGPT Pro?"

Gusarich — Mon, 28 Jul 2025 04:22:26 +0000

Get both ChatGPT Pro and Claude Max if you can. Otherwise, it's a tough choice and depends on how much of your AI time is spent coding versus chatting. If you mostly code, go with Claude Max; otherwise, I'd prefer ChatGPT Pro + Claude Pro.

New comment by Gusarich in "Ask HN: What are you working on? (July 2025)"

Gusarich — Mon, 28 Jul 2025 04:14:31 +0000

A configurable and extensible orchestrator for Claude Code (and other agents). Turns out it can be effectively used not only for coding, but also for reviews, testing, and other tasks. https://github.com/Gusarich/orchestrator

New comment by Gusarich in "Ask HN: Best Practices for Writing Tests?"

Gusarich — Fri, 18 Jul 2025 11:24:21 +0000

That's too generic a question. There are many different techniques, frameworks, and approaches, so there's no single universal answer. But assuming you're asking about testing general-purpose software without specific requirements, I'd recommend the following approach:

1. Basic tests that ensure your software runs without crashing or breaking.

2. Specific test cases for various inputs, settings, and invalid scenarios.

3. Edge-case tests covering extreme inputs or unusual combinations.

4. If your software's behavior is predictable or you can easily verify correctness, consider randomized testing to validate thousands of random inputs automatically.

New comment by Gusarich in "4o Image Generation"

Gusarich — Wed, 26 Mar 2025 10:03:57 +0000

It produces amazing results for me! But the wow effect would have been greater if they had released it a few months ago.