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•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.

Karen Hao: AI creating a DESPERATE BASE OF WORKERS with no full-time employment

https://www.youtube.com/watch?v=E4Zd9ZXjkao
1•chvid•59s ago•0 comments

SnapIndex turns browser screenshots into searchable notes

https://www.snapindex.tools/
1•walkingsardine•2m ago•0 comments

Mathematical Patterns in African American HAIRSTYLEs

https://math.buffalo.edu/mad/special/gilmer-gloria_HAIRSTYLES.html
2•marysminefnuf•7m ago•0 comments

A Place for Free Speach

https://openvoice.club
1•DevBookDigital•8m ago•1 comments

Structured LLM Learning Path, from Zero to AI Researcher, 8-Phase Curriculum

https://github.com/barvhaim/llm-learning-path
1•bignet•8m ago•0 comments

Apiffuf (ffuf for APIs) – fuzzer for API hostsnames and endpoints

https://github.com/jsmonhq/apiffuf
1•rashahacks•13m ago•0 comments

APKPure is distributing a malicious copy of Telegram

https://xcancel.com/EricParker/status/2058411298195661221
2•csmantle•13m ago•0 comments

From source code 2 LLM constraints:a semantic extractor for Python, SwiftUI, Lua

https://github.com/whitecell-dev/Semantic-Extractor/tree/main
1•MaykonMan•14m ago•0 comments

Greg Brockman: Inside the 72 Hours That Almost Killed OpenAI

https://fs.blog/knowledge-project-podcast/greg-brockman/
1•prakashqwerty•16m ago•0 comments

Computer Science Achievement and Writing Skills Predict Vibe Coding Proficiency

https://dl.acm.org/doi/10.1145/3772318.3791666
2•galaxyLogic•16m ago•1 comments

Tinkersynth

https://www.tinkersynth.com/
1•saikatsg•18m ago•0 comments

A maintainability ratchet for AI-assisted Python

https://kayhan.dev/posts/014-letting-agents-write-code-without-ratcheting-up-risk/
1•keynha•23m ago•0 comments

Why Your Docker Container Is 1.2GB When It Should Be 80MB

https://sandeepbansod.medium.com/why-your-docker-container-is-1-2gb-when-it-should-be-80mb-7b443a...
2•thunderbong•27m ago•1 comments

Show HN: TapToyPia

https://memalign.github.io/m/taptoypia/index.html
1•memalign•28m ago•0 comments

Why AI Hardware Is a Chip Layer Problem

https://www.easelinktech.com/why-every-electronic-product-may-need-to-be-rebuilt-for-on-device-ai...
1•EnXu•30m ago•0 comments

Hershey is a textual vector font format

https://wiki.xxiivv.com/site/hershey.html
1•birdculture•30m ago•0 comments

All Lean Books and Where to Find Them

https://lakesare.brick.do/all-lean-books-and-where-to-find-them-x2nYwjM3AwBQ
1•atomicnature•32m ago•0 comments

Instantly understand why any stock is moving

https://whystockmove.com
1•crixusbaba•37m ago•0 comments

InboxFlow Agent – Check your email campaigns before sending

https://github.com/selimsevim/QA-testing-agent
1•selimsevim•39m ago•0 comments

Claude hack – Don't waste you token where it's not needed

https://github.com/justinjamesmathew/tokenmax-mcp
1•justin1006•43m ago•0 comments

Ask HN: I mapped 6,494 AI engines into a taxonomy – anyone else tried this?

1•Developer_H•43m ago•0 comments

1,700 new protein found in the "dark proteome" of previously overlooked DNA

https://pubmed.ncbi.nlm.nih.gov/42092140/
2•janandonly•44m ago•0 comments

Coding agents are giving everyone decision fatigue

https://stackoverflow.blog/2026/05/21/coding-agents-are-giving-everyone-decision-fatigue/
3•vinhnx•51m ago•0 comments

It's Time to Walk

https://www.inferterra.com/the-new-workspace-a-first-principles-exploration-of-dictation-agents-a...
1•vinhnx•52m ago•0 comments

Prof or Hobo?

https://www.proforhobo.com/
1•Paracompact•55m ago•0 comments

The Four Motivators (2016)

https://apenwarr.ca/log/20161213
2•downbad_•1h ago•0 comments

Sleep and Mortality: A Population-Based 22-Year Follow-Up Study (2007)

https://fermatslibrary.com/s/sleep-and-mortality-a-population-based-22-year-follow-up-study
2•downbad_•1h ago•1 comments

Local Root Privilege Escalation and Credential Disclosure in the Linux Kernel

https://blog.qualys.com/vulnerabilities-threat-research/2026/05/20/cve-2026-46333-local-root-priv...
1•signa11•1h ago•0 comments

China behind in LLM race but it can still win in AI, ex-Tencent AI lead says

https://www.scmp.com/tech/big-tech/article/3354394/china-losing-llm-race-it-can-still-win-ai-ex-t...
1•01-_-•1h ago•0 comments

Newsom signs order aimed at tackling AI job displacement

https://thehill.com/policy/technology/5889582-california-ai-job-losses/
1•01-_-•1h ago•0 comments