frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Show HN: Locus – AI agents that ship your code

https://github.com/asgarovf/locusai
1•asgarovf•1m ago•0 comments

Show HN: Arc Raiders companion – Quest tracker,maps,crafting and bots weaknesses

https://arcraiders.website
1•wainguo•3m ago•0 comments

Dishonest People Self-Select into Public Service (In China) [pdf]

http://www.sdwang.org/uploads/4/4/8/5/44856715/bad_apples_1119.pdf
2•marojejian•4m ago•1 comments

New GoDaddy Terms of Service: we no longer serve "consumers"

https://domainnamewire.com/2026/02/19/new-godaddy-terms-of-service-we-no-longer-serve-consumers/
1•bhartzer•5m ago•0 comments

Violent and non-violent death tolls for the Gaza conflict: new primary evidence

https://www.thelancet.com/journals/langlo/article/PIIS2214-109X(25)00522-4/fulltext
1•tareqak•5m ago•1 comments

Add dynamic stats section to home page

1•nishiohiroshi•5m ago•0 comments

Why Most AI Agent Directories Are Basically Useless

https://www.agentrank.tech/blog/why-most-ai-agent-directories-suck
1•hughmcinnis•6m ago•0 comments

A lightweight Windows tool for surfacing unusual system activity

1•EricAUS•7m ago•0 comments

Show HN: Claude-Nonstop – Auto Account Switching and Slack Remote in Claude Code

https://github.com/rchaz/claude-nonstop
1•rchaz•8m ago•1 comments

AMC Theatres Will Refuse to Screen AI Short Film After Online Uproar

https://www.hollywoodreporter.com/movies/movie-news/ai-short-movie-amc-theaters-1236509143/
1•mikhael•9m ago•0 comments

Dishonest People Self-Select into Public Service (In China) [pdf]

https://bfi.uchicago.edu/wp-content/uploads/2026/01/BFI_WP_2026-17.pdf
2•marojejian•9m ago•1 comments

Gemini Pro 3.1's Sage Take on HN and YC

https://gist.github.com/crisdosaygo/1df53af43874192a516a257f5bf06b93
2•keepamovin•13m ago•1 comments

Content Security Policy (CSP)

https://developer.mozilla.org/en-US/docs/Web/HTTP/Guides/CSP
1•doener•17m ago•0 comments

Show HN: Building an e-commerce MVP in in 5 prompts using rdd

https://medium.com/@leechanchai/i-built-an-e-commerce-mvp-in-5-prompts-with-rdd-requirement-drive...
1•cclth•18m ago•1 comments

Fighting the Intel Management Engine with a Cheap System76 Laptop

https://matthewsigmond.com/posts/blog/galago/
2•matthew28845•19m ago•0 comments

OWASP Top Ten Web Application Security Risks

https://owasp.org/www-project-top-ten/
1•doener•22m ago•0 comments

Ask HN: Does treating Inflation as a "Quantization Snap" resolve slow-roll?

1•aplowe•25m ago•0 comments

Grocy "Home Management"

https://github.com/grocy/grocy
2•sourcegrift•26m ago•1 comments

Tracekit: Find what your AI coding agent wastes money on and fix it

https://github.com/0xKoda/tracekit
1•handfuloflight•26m ago•0 comments

Pardoned Binance Founder Hobnobs with Trump Sons, at Mar-a-Lago Crypto Fest

https://www.wsj.com/politics/policy/pardoned-binance-founder-hobnobs-with-trump-sons-administrati...
5•Betelbuddy•28m ago•0 comments

Quod 64kb FPS and deep dive video

https://daivuk.itch.io/quod
1•tonym128•33m ago•2 comments

Stripe closed our account over upsell transaction architecture, not fraud

2•JohannesSchip•34m ago•0 comments

PEP 747 – Annotating Type Forms

https://peps.python.org/pep-0747/
2•azhenley•34m ago•0 comments

Ring's Founder Knows You Hated That Super Bowl Ad

https://www.nytimes.com/2026/02/19/business/ring-super-bowl-ad-privacy.html
1•jhonovich•35m ago•0 comments

Ask HN: Why does it feel like qualifications are irrelevant to hirers?

1•hdhdhsjsbdh•36m ago•2 comments

Use digests, not tags, in your Dockerfiles

https://interrupt.sh/blog/dockerfile-tags/
2•arwt•38m ago•0 comments

Show HN: Mercury CLI – CLI to Connect to Mercury Bank

https://www.npmjs.com/package/mercury-cli
1•ex3ndr•39m ago•0 comments

HHS Releases 6 Years of Medicaid Claims Data ($1T)

https://opendata.hhs.gov/
1•dnw•40m ago•0 comments

Brewing possibilities: Using caffeine to edit gene expression

https://phys.org/news/2026-01-brewing-possibilities-caffeine-gene.html
1•PaulHoule•41m ago•0 comments

Optimism as Resilience and Resistance

https://gathernomoss.substack.com/p/optimism-as-resilience-and-resistance
1•insidiouspaul•42m ago•0 comments
Open in hackernews

My 600 Hours with AI Coding Assistants: A Practical Comparison

2•bv_dev•9mo ago
After spending over 600 hours using various AI coding assistants over the past 3 months, I wanted to share my experience for those navigating this rapidly evolving landscape.

What I Mean by "Agentic Mode" First, to clarify: by "agentic mode," I'm referring to the assistant's ability to understand project context, reason through multi-step problems, and autonomously make coherent code changes across files without constant hand-holding. True agency means the tool can maintain context across interactions and execute on high-level directions.

The Current Landscape (May 2025) Augment Code - Current go-to tool despite higher costs

Strengths: Maintains context remarkably well across complex refactors; actually understands project structure; can implement feature requests that span multiple files Weaknesses: More expensive than alternatives ($30/month vs $20 for others); occasional hallucinations when venturing outside codebase context Best for: Complex refactoring tasks and implementing features that span multiple files

Windsurf - Slightly edges out Cursor for agentic capabilities

Strengths: Better context retention than Cursor; decent file traversal; good understanding of code relationships Weaknesses: Can get quite stuck in their full agentic mode as it starts editing things. While they have removed their flow credits part, it is still painful to watch it go completely out of context. Best for: Mid-size projects where you need moderate autonomy

Cursor - Popular but underwhelming for true agentic work

Strengths: Good IDE integration; clean interface; works reasonably well for single-file tasks. I like the ability to Cmd+K and insert a bulk of code in the middle. Also, I like the @Docs feature to bring latest documentation for popular libraries.

Weaknesses: Context falls apart in agentic mode; often loses track of previous instructions; requires excessive prompting Best for: Single-file optimizations and modifications, but not complex cross-file tasks

Claude Code - Declining quality since public beta

Strengths: Used to have superior reasoning and contextual understanding 3 months ago Weaknesses: Super expensive (like always), but recent updates have significantly degraded agentic capabilities; now requires much more hand-holding than before as it goes compleltely off base. Best for: Simple tasks that don't require deep contextual understanding Note: Most disappointing decline in quality - was previously much more capable. I spent $500 in Feb-Mar and thought it was worth.

Cline, Roo, and Aider - Conceptually interesting but practically limited

Strengths: Cline has good terminal integration; Roo offers interesting visualization; Aider has straightforward CLI Weaknesses: All three struggle with maintaining context; limited understanding of project structure; frequent need to repeat instructions Best for: Very simple, isolated coding tasks or experiments

Real-world Performance Differences The gap between these tools becomes most apparent when trying to implement complex features. For example, when asked to "add user authentication with email verification to my Express app":

Augment Code: Identified relevant files, added middleware, routes, and email service integration, then explained how the pieces fit together Windsurf/Cursor: Added authentication to single files I pointed at but needed explicit instructions for each additional component Others: Generally required file-by-file guidance with frequent context reminders

Conclusion If budget isn't a concern, Augment Code currently offers the most truly agentic experience, but still has a long way to go. For more budget-conscious developers, Windsurf slightly edges out Cursor for agentic capabilities, though both still require significant guidance for complex tasks.

Comments

SoMomentary•9mo ago
I'm always surprised by people sleeping on GitHub Copilot. Is this because people truly don't find any value in it?
bv_dev•9mo ago
I have used Github copilot since their beta release in 2023 and I don't find it anywhere near good these days. Automplete was good, but the industry has moved way beyond 2025. Copilot is slightly worse than Cursor which is itself a pretty average tool now. If you use truly agentic code generation, you won't be able to go back to Github Copilot.
SoMomentary•9mo ago
You don't consider copilots agent mode to be agentic? I've had some pretty great results with agent mode + mcp to have it check it's own work.