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•6mo 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•6mo 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•6mo 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•6mo 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.

AI is hallucinating its way into research

https://thelibre.news/why-is-science-full-of-ai/
1•speckx•2m ago•0 comments

Building an AI-Native Engineering Team

https://developers.openai.com/codex/guides/build-ai-native-engineering-team/
1•gmays•2m ago•0 comments

Google Production Kernel

https://github.com/googleprodkernel
1•doener•2m ago•0 comments

I'm Done with FlutterFlow [video]

https://www.youtube.com/watch?v=8GkOX0i00Mg
1•gozzoo•11m ago•0 comments

The Federated Trust Fabric for the Agent-Native Era

https://www.keycard.ai/blog/the-federated-trust-fabric-for-the-agent-native-era
1•mooreds•11m ago•0 comments

Ask HN: Did Google know about RLHF(breakthru) only after OpenAI shared

1•elarocks•11m ago•1 comments

Washington state Medicare users could soon have claims denied by AI

https://www.kuow.org/stories/thousands-of-washington-state-medicare-users-could-soon-have-claims-...
1•coloneltcb•12m ago•0 comments

I Set a Trap for a Book-Marketing Scammer

https://rwwgreene.substack.com/p/i-set-a-trap-for-a-book-marketing
1•rznicolet•13m ago•0 comments

POTUS says he will sign 'One Rule' executive order to federalize AI regulation

https://www.foxbusiness.com/politics/trump-says-he-sign-one-rule-executive-order-federalize-ai-re...
1•schmuckonwheels•13m ago•1 comments

MicroTimes Interviews T/Maker's Heidi Roizen and Royal Farros (1994)

https://computeradsfromthepast.substack.com/p/microtimes-interviews-tmakers-heidi
1•rbanffy•13m ago•0 comments

Why Hyundai Has to Recall Only Silver Cars over a Serious Safety Defect

https://www.theautopian.com/why-hyundai-has-to-recall-only-silver-cars-over-a-serious-safety-defect/
3•nreilly•16m ago•1 comments

CSS Wrapped 2025

https://chrome.dev/css-wrapped-2025/
1•zhs•17m ago•1 comments

If Your Electron App Is "Just a Browser Wrapper" You're Doing It Wrong

https://www.jasonscheirer.com/weblog/on-electron/
1•speckx•18m ago•0 comments

Static Linux

https://sta.li/index.html
2•Koshkin•18m ago•0 comments

Europe Is Under Siege

https://www.noahpinion.blog/p/europe-is-under-siege
3•alephnerd•20m ago•0 comments

Spotify Wrapped but for Your Terminal

https://github.com/thegoldenmule/terminal-wrapped
1•benthehutt•21m ago•0 comments

Show HN: DataKit, your all in browser data studio is open source now

https://github.com/Datakitpage/Datakit
1•aminkhorrami•22m ago•0 comments

An open-source, cross-platform Git client

https://github.com/markrai/gitpow
1•nhatcher•22m ago•0 comments

Freelancers are losing ~30% of their time and revenue to tool fragmentation

1•Abhijeetp_Singh•26m ago•0 comments

GitHub pull request suggestions escape repository context

1•mrdziuban•28m ago•0 comments

Congress warned that NASA's current plan for Artemis "cannot work"

https://arstechnica.com/space/2025/12/congress-told-there-needs-to-be-consequences-for-nasa-delay...
1•nobody9999•28m ago•1 comments

Generative AI use continues to rise

https://birchtree.me/blog/generative-ai-use-continues-to-rise/
2•speckx•28m ago•0 comments

Defunct Pennsylvania oil and gas wells may leak methane and metals into water

https://phys.org/news/2025-11-defunct-pennsylvania-oil-gas-wells.html
1•PaulHoule•29m ago•0 comments

Market fears mount over private credit inflating the AI bubble

https://www.capitalbrief.com/article/market-fears-mount-over-private-credit-inflating-the-ai-bubb...
3•zerosizedweasle•30m ago•0 comments

A Double Agent in Tehran

https://archive.ph/SgZMm
1•rtmorgan•31m ago•0 comments

Show HN: OpsOrch – Unified API for Incidents, Logs, Metrics, and Tickets

https://www.opsorch.com/
6•yusufaytas•31m ago•0 comments

JuliaGrid: Julia Power System Simulation

https://mcosovic.github.io/JuliaGrid.jl/stable/
1•shallow-mind•32m ago•0 comments

ReductrAI – An AI brain for your infrastructure you can query in plain English

https://reductrai.com/
1•hermosillo_21•32m ago•1 comments

Meta acquires AI device startup Limitless

https://techcrunch.com/2025/12/05/meta-acquires-ai-device-startup-limitless/
2•chermanowicz•32m ago•1 comments

DuckPond: Building a Self-Hosted Data Warehouse with DuckDB, FastAPI, and React

https://datamethods.substack.com/p/duckpond-building-a-self-hosted-multi
1•zekrom•33m ago•0 comments