Podría no ser realmente un empleado de Niantic, pero hasta que lo desmientan... Tiene pinta de ser auténtico. Está en inglés, obviamente:
I worked at Niantic full-time for more than a year
Pros
- Competitive initial compensation compared to large tech companies like Google, if you believe in their options' value.
- Interview process is easier than large tech companies with similar pay.
- 10 year exercise window for the options, so no fear to leave any time after the first vesting.
- Good work-life balance (at least for the platform teams). Playing Niantic games during work hours is somewhat encouraged.
- Most people are friendly.
Cons
- They claim to have an open culture, but don't accept negative comments even when things suck. As a platform engineer who just wanted the game/company to be better, I got punished by echoing Pokemon Go players' complaints which had been there for more than a year. E.g. players with close to 200 friends have to spend 20 minutes every day just to send gifts, which is repetitive and produces no value to the company; exclusive moves make the same pokemon caught/evolved earlier useless, even if it was a Mewtwo caught one year ago by spending money. (They're fixing some of those issues, but much later.) If people don’t agree something is negative, how can anyone start fixing it and making it positive?
- Sometimes even constructive suggestions are not acceptable. I suggested a reusable dataflow (map reduce) to fix multiple player-facing issues like the missing shiny Entei/Suicune for the initial 22 hours (Sep 2, 2019), missing Psystrike for 10 minutes after UltraUnlock week 3 (Sep 23, 2019), and missing Shadow Ball from EX raids during New Taipei City safari zone (Oct 5, 2019). Based on my estimation this would only require a couple eng days of one-time initial work and one eng hour for each individual fix. But an executive basically asked me to shut up and said "It isn't helping to suggest solutions only." (I had volunteered to help on 2 other larger Pokemon Go tasks and was willing to help this one as well, but at the very least I need permissions.)
- Heavy politics in the calibration process. My manager is the same level as me and cannot join my calibration meeting. My performance score of 2019 was reduced from 4.0 (Strongly Exceeds Expectation) to 2.5 (Inconsistent Performance, 3.0 is Meets Expectation), mainly because of the aforementioned executive's complaint on my "behavior". No one else in the calibration meeting had worked with me closely or had 1:1 with me before that, so I had basically no way to affect my final score. The person who could but didn't consider any of my positive "behavior" during the calibration meeting was later "shocked" when I wanted to leave and spent more than an hour trying to convince me to stay. What gives? You gave me a 2.5 when my skill is worth 4.0, of course I will leave. Why did I have to shut up for 2 months until I got some vested options and a job offer just to discuss this issue equally? Why did you want to keep someone who missed your expectation any way?
- Though most employees play it, few people on the Pokemon Go team really know the game well (or the people who know always have other “high priority” stuff to worry about instead of the actual player issues). Before the mass clockblocking of central/mountain time zones’ EX passes on Apr 10, 2019, Silphroad (reddit) already had a post predicting that with 200+ upvotes. But no one on the team did anything. So after that I had to notify them proactively to avoid similar issues twice, including the conflict with Regigigas ticketed event on Nov 2, 2019. (It was simple to postpone the invites before they were sent, but no one else did anything until I warned them in the last hour.) There were other cases where I was able to improve things simply because I understand both the game and some of their tech stack. (My suggestion resulted in the first and simple way to monitor shiny pokemon caught by players, to help prevent missing shiny. I'm glad that they did accept some of my suggestions.)
- Poor prioritization on new features vs. fixing existing issues/bugs. AR buddy multi-player took a lot of engineering effort but do players really use it? Why did some long-lasting issues only get prioritized and fixed 2+ years later? To what extent do they consider players’ voices to be large enough? Why did the non-spawn issue on Salamis island get prioritized immediately after some press coverage, but other posts with 1K+ upvotes just got ignored?
- Poor feature design process which changes spec a lot, costs additional engineering effort and introduces unnecessary issues. The 20-minute gifting issue was because friendship level was supposed to be asymmetrical initially, so a proper fix requires a lot of changes.
- Mediocre engineers. Because the interviews are simpler, the average engineers are not as good as big tech companies’ engineers (Google or Facebook). Engineers’ levels also tend to be inflated. Some “Senior Software Engineers” struggled with designing small systems. Some “Staff Software Engineers” didn’t know much about database transactions even after using them for a while. The same scalability issue on database indexes fixed one week after Pokemon Go launch (and covered in a tech talk) happened again after the recent AR buddy launch. If you play Niantic games, you probably experienced enough issues which could be avoided by better engineers. (There are great engineers at Niantic, but far from enough.)
- Though there wasn't another formal valuation/funding round, the real value of the company probably went down because Harry Potter Wizards Unite is far below expectation. IMO WB is the main one to blame, but some people started losing hope of the company as well.
- Check blind. I’ve said a lot, check what other Niantic employees said on blind, especially after Jun 2019 (Harry Potter Wizards Unite launch).
Advice to Management
- Either stop the false advertisement of the open culture, or start accepting negative comments made by people who just want the company to be better. If something sucks, constructive suggestions alone can not let every one notice the real severity.
- Stop politics and spend more resources on improving players’ trust. Do you really need a strong competitor to start worrying about losing players? (Dragon Quest Walk is already strong enough in Japan.
- Have you really done enough to keep good engineers?
Si es cierto, muchas cosas ya las suponíamos y ahora estarían confirmadas:
- La mayoría de los empleados de Niantic no juega a sus propios juegos.
- Arreglar fallos no está en su lista de prioridades, prefieren meternos la AR por el gaznate, a ver si cuela.
- Leen Reddit, por lo que conocen los fallos que provocan a las pocas horas de lanzar algo nuevo y los fallos que llevan semanas sin solucionarse, pero la mayoría son ignorados... a menos que la prensa se haga eco.
- HPWU no se iba a acercar en éxito a PoGO.
