frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

AI agents attempted exploits on Epoch AI's website during an alignment eval

https://epoch.ai/publications/incident-report-ai-agents-attempted-exploits-on-our-website
1•HowardLien•1m ago•0 comments

Against Supplements

https://worksinprogress.co/issue/against-supplements/
1•surprisetalk•1m ago•0 comments

Ask HN: How likely is Jensen Huang's return as CEO to Nvidia?

1•debrisontrack•2m ago•0 comments

Starter Story: From quitting at $1,700 a month to a sale to HubSpot

https://www.readfounded.com/p/starter-story
1•jcox06•5m ago•0 comments

GPU-Initiated DMA Copy Engines in RCCL and RocSHMEM

https://rocm.blogs.amd.com/software-tools-optimization/gpu-initiated-sdma/README.html
1•ankitg12•5m ago•0 comments

Meta fined $887M in Italy over Android gallery photos, videos on Facebook

https://abcnews.com/Technology/wireStory/meta-fined-887-million-italy-android-gallery-photos-1381...
1•NeuralOracle•6m ago•0 comments

Paint in the Memory of Other Programs

https://citrons.xyz/journal/2026-04-13
1•birdculture•9m ago•0 comments

Show HN: Offline-first grow diary PWA – no account, data stays in the browser

https://apps.plantclue.com/growlog/
1•cannabisdoc•14m ago•0 comments

The Web Sucks

https://suckless.org/sucks/web/
2•zero1009•16m ago•0 comments

I'm making a programming language

https://alfonsocampodonico.com/posts/making-a-programming-language
4•goatley•17m ago•0 comments

One Proton Away

https://konvoy.beehiiv.com/p/one-proton-away
2•HFKG•17m ago•0 comments

Yuval Harari's AI Prediction Includes a Warning of Extinction and More [video]

https://www.youtube.com/watch?v=XPb0GzjeoE4
1•fork-bomber•18m ago•0 comments

Google Search Tests Sign in to Verify Searches Are Human (Not Captcha)

https://www.seroundtable.com/google-search-sign-in-verify-41808.html
1•exploraz•19m ago•1 comments

Advent of Code 2026

https://adventofcode.com/
1•ViktorasJucikas•19m ago•0 comments

Show HN: My developer portfolio with an interactive terminal and live mascot

https://www.yashnandvana.in
1•yashnandvana•21m ago•0 comments

The Secret ThinkPad Powerbutton Code to Bring Dead Laptops Back to Life (2007)

https://masnick.com/posts/the-secret-thinkpad-powerbutton-code-to-bring-dead-laptops-back-to-life/
2•sammatilda•22m ago•0 comments

UCZ – A "zero-prior" context-mixing compressor

https://github.com/no-nova/UCZ
1•adenson•22m ago•0 comments

Albedo: Long running, efficient homes for your agents

https://albedo.sh/
1•icy•23m ago•0 comments

Container-Based Networking Labs

https://containerlab.dev/
1•pelasaco•24m ago•0 comments

Show HN: Open Logs – Self-hosted log management on your own Cloudflare account

https://github.com/Adithya188288/open-logs
1•adi188288•24m ago•0 comments

Coolest Design

https://coolest.design/
1•heyzsolt•24m ago•0 comments

Implicit state machines hide bugs that line-by-line review approves

https://quickchat.ai/post/implicit-state-machines-hide-bugs-that-line-by-line-review-approves
2•piotrgrudzien•27m ago•1 comments

I built a supplier-rating tool to replace our quality spreadsheet

https://suppliscore.com/
1•maleni•29m ago•0 comments

What a 30 Day Digital Journaling Experiment Can Still Teach Us

https://nathanohren.com/paper-pixels-and-the-private-self-what-a-30-day-digital-journaling-experi...
1•threecells•30m ago•0 comments

Show HN: An interactive map showing biodiversity data across the world

https://app.taxonomap.com/
1•anotherjack•30m ago•0 comments

Rubik's Cube production returns to Hungary more than 50 years after invention

https://www.euronews.com/video/2026/11/14/rubiks-cube-production-returns-to-hungary-more-than-50-...
2•giuliomagnifico•32m ago•0 comments

HN Show: Cubepals - Open-source Minecraft hosting, servers sleep until you join

https://cubepals.com
1•xlmnxp•33m ago•0 comments

Microsoft – more details on the root cause

https://twitter.com/clamanna/status/2122209697226948729
1•hyperman1•33m ago•0 comments

The small African lab sounding the alarm on AI bioweapons

https://restofworld.org/2026/ghana-ai-lab-minohealth-bioweapons-threat-dna-synthesis-regulation/
2•milali•36m ago•0 comments

Running a 6M parameter model on a Game Boy

https://twitter.com/timhwang/status/2122011842549600506
2•MrBuddyCasino•40m ago•0 comments
Open in hackernews

My 600 Hours with AI Coding Assistants: A Practical Comparison

2•bv_dev•1y 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•1y 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•1y 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•1y 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.