frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

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.

Show HN: Posecode inspectable human movement as text

https://www.posecode.org/play
1•brnbrn•39s ago•0 comments

AI Data Center Alert

https://aidatacenterimpact.com/
1•studsnchayns•3m ago•0 comments

Get ready to flee, Americans in ten countries warned

https://www.dailymail.com/news/article-16021741/flee-Americans-nine-countries-US-embassies-Iran-w...
1•Bender•3m ago•0 comments

Charities remain locked out of CAF Bank online accounts

https://www.theregister.com/security/2026/07/31/charities-remain-locked-out-of-caf-bank-online-ac...
1•Bender•6m ago•0 comments

Kun's Pi Agent Config

https://blog.kunchenguid.com/p/kuns-pi-agent-config
1•handfuloflight•10m ago•0 comments

Fermi Paradox

https://en.wikipedia.org/wiki/Fermi_paradox
1•simonebrunozzi•11m ago•0 comments

I made Squirrel game with Flutter and Flame

https://danvilela.com/squirrel-up
1•dmvvilela•12m ago•0 comments

Show HN: Cockpit for you Claude Code agents in Rust

https://episko.dev/
1•evolabs•12m ago•1 comments

GenAI Added to Google Earth

https://www.theatlantic.com/technology/2026/07/google-earth-ai-images/688145/
2•alwa•13m ago•1 comments

People Are Crossing Oceans for a Fayetteville Roller Coaster

https://www.axios.com/local/atlanta/2026/07/24/arieforce-one-closing-fayetteville-fun-spot-americ...
2•Xcelerate•14m ago•0 comments

Rails patches critical Active Storage flaw with RCE potential

https://www.bleepingcomputer.com/news/security/rails-patches-critical-active-storage-flaw-with-rc...
2•sbulaev•14m ago•0 comments

Tell HN: Amazonbot aggressively scraping my website and ignoring robots.txt

1•pera•17m ago•4 comments

Trump Media's new paid data service goes live, faster access to Trump's posts

https://www.cnbc.com/2026/08/01/trump-medias-new-data-service-gives-faster-access-to-trumps-posts...
1•srameshc•18m ago•0 comments

Static Devirtualization of Tencent VM

https://back.engineering/blog/31/07/2026/
2•not_a9•20m ago•0 comments

The Swiss Spaghetti Harvest

https://hoaxes.org/archive/permalink/the_swiss_spaghetti_harvest
1•EndXA•21m ago•0 comments

Signal Structure of the Starlink Ku-Band Downlink (2023) [pdf]

https://radionavlab.ae.utexas.edu/wp-content/uploads/starlink_structure.pdf
1•peter_d_sherman•22m ago•0 comments

A Tech Founder Wanted to Start a New Country. An Actual Country Got in the Way

https://www.wsj.com/tech/a-tech-founder-wanted-to-start-a-new-country-an-actual-country-got-in-th...
1•impish9208•23m ago•1 comments

Chess Engine Dev Community Openly Hostile to AI Assisted Development

https://github.com/adamtwiss/coda/issues/15
2•chesspeoplercra•24m ago•0 comments

What's Up with the Weird Mouths of These Finch Chicks?

https://www.audubon.org/news/whats-weird-mouths-these-finch-chicks
1•thunderbong•25m ago•0 comments

Expanded Genetic Code

https://en.wikipedia.org/wiki/Expanded_genetic_code
1•frozenseven•26m ago•0 comments

Google cancels AI Studio app after 800k preorders

https://twitter.com/GoogleAIStudio/status/2083274575769473092
2•BlueBerry2001•26m ago•0 comments

Show HN: BugDetect AI – AI-powered code debugging assistant

https://bugdetectai.com
1•laxmansubedi•31m ago•0 comments

CISA Alert: Water Sector PLC Targeting

https://censys.com/blog/cisa-alert-water-tower-plc-targeting/
1•speckx•31m ago•0 comments

Toward a Framework for Openness in Foundation Models

https://cacm.acm.org/research/unpacking-open-source-artificial-intelligence-toward-a-framework-fo...
1•visha1v•34m ago•0 comments

Sigstore Cosign signature verifier in 150 lines of C

https://github.com/kern-werk/cosign-verify
2•pdubouilh•34m ago•0 comments

Show HN: Cross-platform AirDrop replacement but over internet using Iroh

https://github.com/chad/iroh-drop
1•chadfowler•35m ago•0 comments

TextGrad: Automatic "Differentiation" via Text

https://github.com/zou-group/textgrad
1•rzk•36m ago•0 comments

Show HN: Bribes.fyi - Update on Know before you go feature

https://bribes.fyi/know-before-you-go
2•neverenderr•41m ago•0 comments

Earth, Love and Internet

https://epic.projector.earth/?extension=love
1•blueturn•45m ago•0 comments

Charges Dropped in Reflecting Pool Case as Prosecutors Say Work Was 'Botched'

https://www.nytimes.com/2026/07/31/us/politics/david-hearn-reflecting-pool-vandalism-charges.html
2•whack•49m ago•0 comments