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•8mo 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•8mo 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•8mo 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•8mo 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.

Are There Enough Engineers for the AI Boom?

https://spectrum.ieee.org/ai-data-centers-engineers-jobs
1•JeanKage•2m ago•0 comments

Show HN: Gh-PR-review – CLI tool for LLMs to create, read, comment PRs

https://github.com/agynio/gh-pr-review
2•NBenkovich•6m ago•0 comments

Unveiling Firestore Pipeline operations – Firestore's powerful new query engine

https://firebase.blog/posts/2026/01/firestore-enterprise-pipeline-operations
1•reimertz•8m ago•0 comments

Show HN: Build AI Agents Declaratively with Terraform

https://github.com/chatbotkit/terraform-provider-chatbotkit
2•_pdp_•8m ago•0 comments

RISC-V is coming along quite speedily: Milk-V Titan Mini-ITX 8-core board

https://www.tomshardware.com/pc-components/cpus/milk-v-titan-mini-ix-board-with-ur-dp1000-process...
1•fork-bomber•8m ago•0 comments

Ask HN: Which skill is best for future?

1•SRMohitkr•9m ago•0 comments

Mandiant releases rainbow table that cracks weak admin password in 12 hours

https://arstechnica.com/security/2026/01/mandiant-releases-rainbow-table-that-cracks-weak-admin-p...
1•thunderbong•10m ago•0 comments

Does Antimatter 'Fall Up'?

https://www.space.com/astronomy/dark-universe/does-antimatter-fall-up
1•AdamN•10m ago•0 comments

The next frontier for public access: building channels of meaning

https://www.aaas.org/news/next-frontier-public-access-building-channels-meaning
1•JeanKage•14m ago•0 comments

Show HN: Agentic Commits – Commit spec for AI agent workflows

https://agentic-commits.deligoz.me
1•deligoez•15m ago•0 comments

Alarm is sacred, must not fail, but iOS 26 is wicked

https://rakhim.exotext.com/alarm-is-sacred-but-ios-26-is-wicked
2•7777777phil•17m ago•0 comments

Palacio Barolo

https://en.wikipedia.org/wiki/Palacio_Barolo
1•f1shy•18m ago•0 comments

JBDev: Jailbreak/TrollStore app development with Xcode

https://github.com/lich4/JBDev
1•lich4•19m ago•0 comments

I built a "Linter" for SaaS features (detects missing billing/auth flows)

https://github.com/SkeneTechnologies/skene-growth
1•teemuki•19m ago•1 comments

Wikipedia: WikiProject AI Cleanup

https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup
5•thinkingemote•19m ago•0 comments

Show HN: HHistAI – Explore History with Artificial Intelligence

https://histai.net/
1•jsxyzb•22m ago•0 comments

Use Mac as a coding agent with no additional server setup

https://getmoshi.app/articles/mac-remote-endless-agent-setup
1•rjyo•23m ago•1 comments

You have three minutes to escape the perpetual underclass

https://geohot.github.io//blog/jekyll/update/2026/01/17/three-minutes
2•7777777phil•26m ago•0 comments

Quadro Tracker

https://en.wikipedia.org/wiki/Quadro_Tracker
1•bschne•27m ago•0 comments

Led by Texas, New Hampshire, U.S. states put Bitcoin on public balance sheet

https://www.cnbc.com/2026/01/17/texas-us-states-budgets-bitcoin-crypto-strategic-reserve.html
1•TMWNN•30m ago•1 comments

Show HN: Gdocs-CLI – Fetch Google Docs as Markdown for AI Coding Agents

https://github.com/famasya/gdocs-cli
1•pacific01•30m ago•0 comments

CVEFinder – Fast CVE lookup with product-level mapping

1•cvefinder•33m ago•1 comments

Ask HN: "Unfortunately, humanity's future is in the hands of [redacted]"

2•fdeage•33m ago•1 comments

The Jolla Phone Proved We've Been Using Smartphones Wrong All Along

https://techglimmer.io/jolla-phone-2026-review-kill-switch-privacy-review/#Jolla
32•mpol•34m ago•18 comments

NASA ID Sy1174304 – O(1) Strategic Proposal: Alzheimer's and Longevity

2•MASTER_shivam•34m ago•0 comments

Last year I eliminated our PTO policy. I called it "unlimited."

https://twitter.com/gothburz/status/2011421898161144192
19•marbartolome•36m ago•9 comments

Castle Engine downloads with bundled FPC for all platforms

https://castle-engine.io/wp/2026/01/18/engine-downloads-with-bundled-fpc-for-all-platforms-castle...
1•mariuz•36m ago•0 comments

2026 Adamuz Train Derailments

https://en.wikipedia.org/wiki/2026_Adamuz_train_derailments
1•Rygian•39m ago•0 comments

Determinate: End-to-End Nix for macOS, WSL, Linux

https://docs.determinate.systems/
1•Lwrless•40m ago•0 comments

Show HN: ChatGPT Projects wasn't enough, so I built my "dream notes app"

https://apps.apple.com/us/app/wiz-note-ai-smart-notes/id6757632086
1•wabiosdev•41m ago•0 comments