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.

Time travellers are using LinkedIn to teach us about artificial intelligence

https://www.ft.com/content/4a4b442c-7ae4-4553-bcab-d744d8d46c79
1•JumpCrisscross•52s ago•0 comments

A Polish architect rotated his house to hide it in the hill

https://www.yankodesign.com/2026/07/30/a-polish-architect-rotated-his-house-to-hide-it-in-the-hill/
1•andrewstetsenko•1m ago•0 comments

Hank Green found the AI problem that YouTube labels can't catch

https://arstechnica.com/ai/2026/08/hank-green-found-the-ai-problem-that-youtube-labels-cant-catch/
1•CharlesW•3m ago•0 comments

Fragplay – Download and play music easily

https://codeberg.org/micro4scopic/fragplay
1•holeinwall•3m ago•1 comments

Pam Bondi Implicated in as Much as $5.5M in Insider Trading

https://cmarmitage.substack.com/p/pam-bondi-implicated-in-as-much-as
3•JumpCrisscross•5m ago•0 comments

Muse Spark 1.2 (Xhigh) Intelligence, Performance and Price Analysis

https://artificialanalysis.ai/models/muse-spark-1-2
1•theanonymousone•6m ago•0 comments

Animals that self-medicate (2014)

https://www.pnas.org/doi/10.1073/pnas.1419966111
1•EndXA•9m ago•0 comments

Call for Ideas: Autonomous Software Experiments on Hera

https://ideas.esa.int/core/servlet/hype/IMT?documentTableId=8527279057944083272&userAction=Browse...
1•Carbonhell•10m ago•0 comments

Mexico's top univ orders 58k students to retake entrance exam:suspected cheating

https://www.theguardian.com/world/2026/aug/05/mexico-top-university-unam-resit-entrance-exam-chea...
1•bookofjoe•10m ago•0 comments

Reddit signals ominous upcoming "changes" for old.reddit.com

https://arstechnica.com/gadgets/2026/08/reddit-signals-ominous-upcoming-changes-for-old-reddit-com/
2•CharlesW•11m ago•1 comments

Show HN: Bifrost: Enterprise MCP Gateway with Built-In Security: OAuth 2.0, RBAC

https://github.com/maximhq/bifrost
1•aanthonymax•12m ago•0 comments

Prime Agent

https://twitter.com/PrimeIntellect/status/2085086999267144083
1•tosh•13m ago•0 comments

OpenStamp – A watermarking method for open-weight LLMs

https://github.com/mb-14/openstamp/
1•mb14•14m ago•0 comments

AgentGuard – fail-closed approval gateway for AI agent tool calls

https://github.com/moon2022alakhali-commits/agentguard
1•yhouseff_dev•14m ago•0 comments

Tl;Dv (Too Lazy; Didn't Validate): 181,874 Meetings Left Wide Open

https://bobdahacker.com/blog/tldv-hack
1•yoelcabo•15m ago•0 comments

The Lyon Amendment to the Lancaster Consensus: Perl's Toolchain Version Support

https://github.com/Perl-Toolchain-Gang/toolchain-site/blob/master/lyon-amendment.md
1•fanf2•15m ago•0 comments

Double trouble for Microsoft as pre-owned software license claims converge

https://www.theregister.com/saas/2026/08/05/double-trouble-for-microsoft-as-pre-owned-software-li...
1•Bender•16m ago•0 comments

AI-Focused Altimeter Sinks 11%, Joins Funds Hit by Brutal July

https://www.bloomberg.com/news/articles/2026-08-05/balyasny-verition-join-hedge-funds-losing-mone...
1•mapping365•16m ago•0 comments

China is building a blue granary

https://www.palladiummag.com/2026/08/04/china-is-building-a-blue-granary/
2•hunglee2•17m ago•0 comments

Creator of Lua: What People Get Wrong About Scripting Languages [video]

https://www.youtube.com/watch?v=jCZnFKk6M9A
1•skibz•18m ago•0 comments

The Bitter Lesson (2019)

http://www.incompleteideas.net/IncIdeas/BitterLesson.html
1•zbentley•21m ago•0 comments

Show HN: Command Code GOAT, $10/month for $70 of credits across 30 models

https://twitter.com/CommandCodeAI/status/2085039006203711649
5•ahmadawais•22m ago•2 comments

Conductance levels in memristors integrated on CMOS – Nature

https://www.nature.com/articles/s41586-023-05759-5
2•bilsbie•22m ago•0 comments

The Lisp Machine: Your Next Step in Computing

https://www.symbolics.biz
1•yacin•25m ago•0 comments

Apple Vision Pro Speeds Up Endoscopic Surgery by Almost 20%

https://www.macrumors.com/2026/08/05/apple-vision-pro-speeds-up-surgery/
3•MBCook•25m ago•1 comments

OpenMouse: Open-source, browser-based app for managing computer mice

https://openmouse.app/
1•morjom•25m ago•1 comments

Tcrf.net served Claude a payload telling it to wipe the working directory

https://github.com/bashalarmistalt/tcrf-ai-agent-payload-report
1•BashAlarmist•26m ago•1 comments

OpenAI agents rebuilt a secret message board after the company shut it down

https://runtimewire.com/article/exclusive-openai-agents-rebuilt-a-secret-message-board-after-the-...
2•ryanmerket•26m ago•0 comments

Visual Prompt Engineering for Video Models

https://visual-prompt-engineering.github.io/
1•gmays•27m ago•0 comments

Cross-Compilation with Wails

https://chriswheeler.dev/posts/cross-compilation-with-wails/
1•auraham•28m ago•0 comments