week

Attention over time

7-day rolling share of analyzed posts/comments mentioning this entity · 90 daily points ending 2026-08-28

Final complete day: 2026-08-28. Attention was 0.5% — this entity appeared in 13 of 2,594 analyzed posts/comments in the seven days ending that day (15 mentions).

Each point is a seven-day rolling window of complete UTC days; a point is comparable only when at least 100 analyzed posts or comments fall inside it. A day with no analyzed records counts as zero; missing coverage is marked and never shown as zero. The series ends at the latest complete analyzed day — it never extends into the current incomplete day. How attention is measured.

Recent mentions

1. 2026-08-28 19:19 UTCComment

   don't sell her the tracker, sell her the thing she already wants: not being the person who has to remember everything. what worked for me with a client who lived in outlook: i kept using the issue tracker on my side and stopped arguing about it. every friday she got one email from me, plain text, three sections — done this week, in progress, waiting on you. the "waiting on you" list had dates on it. that's it, no tool, no login. two things happen. she starts forwarding that email to her own boss because it makes her look organized, and when something slips there's a written record from before it slipped showing it sat waiting on an answer for nine days. after about a month she asked if there was "somewhere she could just look at that list" instead of waiting for friday. that's the invite moment. before that it's just extra homework you're assigning her. the other half: never accept a task that only exists in an email thread. reply with "adding this as issue #— confirming it's higher priority than the checkout bug we agreed on monday, yes?" you're not asking her to use github, you're making her pick, and email is a terrible place to keep track of what you picked. she'll feel that herself pretty fast. and yeah, keep doing the go-between work in the meantime. it's annoying but it's about an hour a week and it's what buys you the leverage. forcing it in week one with a new pm who controls whether you get paid is a fight you win on paper and lose everywhere else.

2. 2026-08-28 12:49 UTCComment

   They were suggested as one of the sites to use on the Odin project. Needless to say within like a week of hosting the project I already get mail of the project being paused. How's anyone supposed to try out the stuff I made? Useless.

3. 2026-08-28 10:26 UTCComment

   Hard to put an exact number on it, but the maintenance burden rarely comes from app count—it's the silent third-party API edge cases that consume your entire week. Case in point: I've been chasing an Airtable API bug for over a week now. The pipeline is architected to execute strictly 2 API calls per build, but Airtable's usage reporting keeps flagging 8–9 calls per run even when raw server logs confirm exactly 2 outgoing requests. When you're managing multiple properties solo, one mystery discrepancy like that will burn 10+ hours of troubleshooting, completely stalling active feature development. If management doesn't see those unexpected integration fires, they just assume you're spending 100% of your time building new stuff.

4. 2026-08-28 00:08 UTCComment

   Everyone’s covered that this is a business, not a website. I’d avoid trying to solve it by finding the correct big number. The doc looks detailed, but it mixes build work with a pile of product decisions nobody has made yet. “Multiple polished themes,” “GDPR-compliant,” “an RSVP or payment must never be lost,” tier rules, guest-photo moderation, payout failures, refunds, what happens when someone changes an RSVP after seating is done. Those aren’t small implementation details, they decide the build. I’d propose a paid discovery phase first. The output is not code. It is: * a cut-down v1 with a single end-to-end event flow * a list of decisions the client owns, with examples and acceptance criteria * the integrations and operational risks, especially payments, notifications, and uploaded photos * milestone scopes, exclusions, and a change process for everything added later * a separate monthly support and maintenance arrangement after launch Then quote the first build milestone, not “the platform.” For example, one theme, event page, RSVP, basic guest list, and transactional email. Leave seating charts, Connect payouts, SMS, photo uploads, analytics, AI edits, and corporate invoicing for later unless the business case says otherwise. Ask for budget directly. If their budget only buys a prototype, that is useful information before you spend weeks turning their wish list into your fixed-price liability.

5. 2026-08-27 19:32 UTCPost

   DO NOT interview for Mimica.ai, truly torturous and LONG take home test and interview process - To name and shame bad companies, I recently interviewed with for a Staff fullstack developer role and it was a diabolical and unrealistic process and I was shocked to see that actual developers designed those process. 1. Round 1, HR 30 mins 2. Immediately I was given a take home test. It was react.js focused and it was ridiculously complex and long for an interview process and coding challenge. Challenge: Consume their tree API and create an infinitely long and infinitely nested tree in the browser with editable leafs whose change of status should cause the immediate parents status recursively *(and the parent should take into account all its direct children's)* based on number of business criteria's and different colours. The acceptance criteria had many scenarios like below: * address all requirements listed above * performance matters - the frontend should be able to maintain a good performance even if the **amount of data** from the backend **increases exponentially** * UI: it should look exactly like the provided screenshots * handle API errors gracefully * make correct use of Redux, following recommended storage patterns * make correct use of React and its change detection function * correctly implement the logic to compute automatable statuses * make use of code design and architecture best practices * follow JS conventions for variable, function, and component names * Document design decisions, architecture, if you had more time what would you do and so on And they told that it would take around 4 hours. I immediately told them that building this from scratch and satisfy every criteria along with documentation is not a 4 hours job and it would take significantly longer to come up with something they will like. But I stupidly did it anyways because they had good glassdoor reviews. It took me several fulltime days to do it in a way which was truly scalable with everything they asked. I submitted the test (7 day cutoff) and they took almost 10 days to get back with a feedback that I scored 100% on the test. 3. The next round was 1 hour live coding *(no AI/LLM allowed which was fine with me)* where i present my solution and then in last 40 mins they ask me to extend the solution. The interviewers were not welcoming and were going through motions like a meat grinder. In last 40 mins they said they expect me to solve 3 extensions. they pasted 2 page requirement for each challenge live which I had to read, understand, ask questions and then implement it all within 40 minutes. Now 3 extensions in such complex code in 40 mins is a lot but I had no choice as I was already there. The first one was done in less than 10 mins and to that they said you architected in a way which kinda alreadysolves this problem and I was left thinking "*so it should be a good thing right, why are they talking as if I commited a crime instead of praising my architecture?*". They seemed visibly unhappy that I completed it effortlessly. Then they gave the second 2 page question and it was not an extension but a complete 180 degree change of requirements which required a complete refactor for the entire core multi day exercise work that I did. They literally changed how all the parents should react and calculate based on all decendents and not just direct children's with a complete change of business criterias which itself took a while to understand and clarify from the 2 page requirements. I told them this is basically a full refactor from the original challenge and that would change the whole core piece touching pretty much everything and expected me to complete in remaining 30 mins. It was pure torture because the whole assignment changed but now I had to do it in remaining 20 mins. I did as much as i could and we ran out of time and I couldn't even get to the last question. Honestly what a wasted and unrealistic interview process to ask a candidate to do a full refactor but instead of few day/several hours, now do it in 20 mins and it tests nothing. At the end of the interview they told me they will share the feedback sometime next week which means another 1 week delay and the initial coding test also took them 10 days to get back to me. If this goes through then there would be a ridiculously unrealistic systems design where they would ask me to design a really really complex software in 50 minutes and if that goes through (and another 10+ days to get the feedback) a chat with founders and if that goes through then the process ends. After going through this unrealistic experience and already losing a month, I shot them the email that I am withdrawing from the process and don't have energy to go through it and potentially spend another month or 20 days in limbo if this round even goes through. Companies truly treat candidates like dirt, especially in the current market and have no respect for candidates time. They think candidate are ONLY interviewing for them and will spend whole week of unpaid work just to do coding exercise for them. And companies won't even bother checking for 10 days and give a patchy review in the end. And what's up with unrealistic interview process. 3 coding "refactors" masquerading as "innocent extensions" in 40 mins. Really? Do those engineers themselves code an entire massive refactor from groundsup in 20 mins? how can they be this delusional. Few years ago I got a coding exercise to create an entire bicycle sharing app with a realtime API and do FE/BE/Unit/Integration tests, documentation, deployed on AWS and bonus "scalable to millions of users". Time, maximum 6 hours. Truly delusional. I laughed and noped out. But i was shocked that so many candidates are doing those because they truly need a job. I just wanted to create this post to share my experience. Already have written glassdoor review for them but wanted to spark a conversation here as well because this is not a topic talked about much. So, what are your experiences of bad companies and what crazy things have they asked you to build for free and their unrealistic expectations. It is time to name and shame companies so that they stop taking unpaid free work from developers under the name of "interview".