frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

My 600 Hours with AI Coding Assistants: A Practical Comparison

2•bv_dev•7mo 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•7mo 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•7mo 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•7mo 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.

The Revolution of Rising Expectations

https://thezvi.wordpress.com/2025/12/22/the-revolution-of-rising-expectations/
1•blfr•1m ago•0 comments

Show HN: JSON-Healer: Repair and Validate Broken JSON from LLM Outputs

1•freakynit•2m ago•0 comments

Summary of the USA federal government's zero-trust memo

https://www.bastionzero.com/blog/i-read-the-federal-governments-zero-trust-memo-so-you-dont-have-to
1•fanf2•3m ago•0 comments

Linux kernel community discussion on ML/LLM tools in kernel development

https://lwn.net/Articles/1050355/
1•todsacerdoti•6m ago•0 comments

Late last night, I received a final Christmas gift

https://twitter.com/PalmerLuckey/status/2004638982915981652
1•keepamovin•6m ago•0 comments

Show HN: I built a secure "Linktree" for IBANs and Crypto addresses

https://iban.bio
2•kadiran•8m ago•0 comments

Ask HN: Do you believe any LLM's pass the Turing test? How?

1•oldputput•8m ago•0 comments

VP of Sales

https://unicursaltalent.com/
1•paulwesleydavis•11m ago•0 comments

Sequence Toys: Train a language model in the browser with WebGPU

https://sequence.toys
1•handfuloflight•13m ago•0 comments

AI village, watch frontier AIs interact with each other and the world

https://theaidigest.org/village
1•Alifatisk•19m ago•1 comments

Show HN: Bibrof AI – Bulk Image Background Remover Offline

https://bibrof.lislip.com/
1•vinayak18696•21m ago•0 comments

Package Forge, Improving Package Management for *Unix Systems

https://github.com/pkgforge
1•rhim•22m ago•0 comments

Own a Graph

https://staysaasy.com/strategy/2025/11/25/own-a-graph.html
1•thisismytest•23m ago•0 comments

Concert cancelled after AI wrongly accuses artist of being sex offender

https://www.cbc.ca/news/entertainment/ashley-macisaac-ai-accusation-9.7026786
2•adam-p•23m ago•0 comments

RoboCop – Breaking the Law. H0ffman Cracks RoboCop Arcade from DataEast

https://hoffman.home.blog/2025/12/26/robocop-breaking-the-law/
1•birdculture•24m ago•0 comments

Geely's SiEngine Surpasses Global ADAS Standards with New 7nm SoCs

https://robotdyn.com/geelys-siengine-surpasses-global-adas-standards-with-new-7nm-socs/
1•teleforce•28m ago•0 comments

Are Private Jets Less Safe Than Flying Commercial?

https://www.vanityfair.com/style/story/private-jet-plane-crashes
1•rbanffy•28m ago•0 comments

Free and Open Sourced Prompt Generator for Image and Software Projects

https://prompy.xyz
1•niyoseris•29m ago•1 comments

Workmux: Git worktrees and tmux for parallel AI agents

https://raine.dev/blog/introduction-to-workmux/
1•rane•30m ago•0 comments

Serde (Rust Serialization Deserialization)

https://serde.rs/
1•mpweiher•33m ago•0 comments

Ask HN: How do your teams keep docs and code in sync?

1•iampotatoman92•35m ago•0 comments

Create the next season of Emily in Paris

https://claude.ai/public/artifacts/9a29a330-fcca-4373-8c1b-904f84e2a4e9
1•valkrieco•36m ago•0 comments

Ask HN: Is there a tracker for OOS internationalization requirements?

1•devdp430•37m ago•0 comments

Show HN: The bedtime, another little bedside clock

https://www.stavros.io/posts/i-made-another-little-bedside-clock/
1•stavros•37m ago•0 comments

How Much Would It Cost the U.S. to "Buy" Greenland?

https://www.thepricer.org/how-much-would-it-cost-the-u-s-to-buy-greenland/
1•canitellyou•39m ago•2 comments

Show HN: I spent 3 months building an AI trading bot using DRL like AlphaGo

https://github.com/zero-was-here/tradingbot
1•solosquad•40m ago•0 comments

3D-Printable High-Strength Al Alloys via Rapid Solidification

https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202509507
1•bookofjoe•41m ago•0 comments

0day unauthenticated RCE affecting 70k devices on the internet found by AI

https://pwn.ai/blog/cve-2025-54322-zeroday-unauthenticated-root-rce-affecting-70-000-hosts
1•shinkurt•44m ago•0 comments

A Geometric Approach to 5D Superfluid Stabilization

1•j_wood•45m ago•1 comments

Digital Signals Theory

https://brianmcfee.net/dstbook-site/index.html
1•phoreverpheebs•47m ago•0 comments