惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

aimingoo的专栏
aimingoo的专栏
Jina AI
Jina AI
WordPress大学
WordPress大学
Recent Announcements
Recent Announcements
G
Google Developers Blog
I
InfoQ
H
Hackread – Cybersecurity News, Data Breaches, AI and More
Google DeepMind News
Google DeepMind News
P
Proofpoint News Feed
MyScale Blog
MyScale Blog
M
MIT News - Artificial intelligence
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
C
Check Point Blog
J
Java Code Geeks
T
Tailwind CSS Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Microsoft Security Blog
Microsoft Security Blog
MongoDB | Blog
MongoDB | Blog
V
Visual Studio Blog
人人都是产品经理
人人都是产品经理
量子位
A
About on SuperTechFans
D
DataBreaches.Net
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders
I built a bookkeeping app for UK sole traders as a new de...
JamesQP · 2026-06-14 · via Hacker News - Newest: "AI"

About a month ago I started building QuarterPerfect — a bookkeeping tool aimed at UK sole traders and landlords, specifically designed around the MTD ITSA changes coming into effect this year (first quarterly deadline: 7 August).

A bit of background on me firstly.

I'm not a senior developer. My previous project was a small roster app that converted an XLSX shift rota into a calendar with annual leave support — mostly built for me and a few colleagues at work. Nothing with real users, nothing with payments or compliance requirements. QuarterPerfect is a significant step up from that.

QuarterPerfect offers users bank statement import (CSV and PDF), transaction categorization with receipt matching, merchant rules for auto-categorizing recurring transactions, and a read-only accountant share link so clients can hand off clean records without chasing. It's web-first with a mobile app in testing.

The technical side was challenging but learnable. The harder thing was building in a domain I have no real background in. The HMRC/MTD space has a lot of nuance, treatment of different income types, what "quarterly updates" actually mean vs the annual return, what accountants need vs what HMRC needs. Getting that wrong isn't a UX problem, it's a liability problem.

The second hard thing was learning to work effectively with AI assistance. I used Claude and Perplexity heavily throughout. Early on I'd accept output too uncritically and build on shaky foundations. The shift that helped most was treating it less like a code generator and more like a pair programmer, the decisions still have to be mine, and I have to understand what I'm shipping well enough to defend it. Honestly though, one really good work flow was to give Perplexity my ideas and have it generate detailed prompts for Claude, I would then post results back to Perplexity. A LOT of copy and pasting but results were spot on. I also had them both agree on a project management structure, which REALLY helped keep things tidy.

It's still in testing, with no public users yet. The product works end-to-end but I'm not comfortable pushing it to strangers until I'm more confident in the compliance side. Also I want the app to be native to iOS and Google Play store, the latter requires me to have a minimum of 12 testers using the app in some capacity for 14 days, I haven't reached the minimum amount yet, even after asking for help from friends on social media, quite deflating to be honest!

Has anyone else built in a regulated or compliance-heavy space without vast expertise? How did you handle the gap between "this works" and "this is correct enough to trust with someone's tax records" for example.

Has anyone else found the AI-assisted development workflow genuinely helped them create a working/useful website/app, what has been your experience?

I'm not looking to recruit users from this post, the product requires signup to evaluate properly anyway. More curious whether anyone's been in a similar position and what they did to get it over the line.

Thanks,

James