jdwyah

12 ms·

121.
by jdwyah 5y ago
Lazy Q, but do you know if there is a reason that the votes need to be whole numbers? If I allocate 7 credits to something is there any reason not to just count that as 2.65 votes? (And by "any reason" I guess I mean "any rea
122.
by jdwyah 5y ago
That UI really is good. The large number of votes (100) and issues (10) seems to reduce the issue around not being able to "spend" all your votes as well. When I've only got 25 it feels particularly wasteful to not be able to
123.
by jdwyah 5y ago
Congrats! The mutual agreement part is fun stuff. And yay for overkill software! I 100% used my ForceRank.it tool to try to align on names. For us we wanted 2 middle names so there was real combinatorial explosion ;)
124.
by jdwyah 5y ago
Monopoly simulator is a fun little project. I had a very strange reason to build one once upon a time because a chef was doing a monopoly themed dinner and needed to know how much food for each square. http://blog.jdwyah.com&#x2F
125.
by jdwyah 5y ago
Every Product Manager: “I dunno, ever since they started saying React, everything takes 3x longer to ship and we have 2 teams on very challenging technical projects to enable performance that doesn’t suck”.
126.
by jdwyah 7y ago
This is great. In a mildly amusing side-note, in order to fill out my companies profile, my (barely profitable) side project is a perfect fit: https://forcerank.it lets you have your team rank all the values so you can see what
127.
by jdwyah 8y ago
Being able to change log levels on the fly with specifics granularity is a game changer imo https://www.prefab.cloud/documentation/logging Otherwise your logs are always either too verbose or not there when you need th
128.
by jdwyah 8y ago
Oh, link to a description https://engineering.ezcater.com/ezexperiments-ab-testing-at-...
129.
by jdwyah 8y ago
Add me to the list of former Segment users. A pricing change made them want to bump our bill from $199 to $10999. I would explain to them that this was insane. They’d grandfather me in for a quarter. But every quarter I’d get new urgent ema
130.
Adding Feature Flags Service to a Rails App [video](blog.prefab.cloud)
1 points by jdwyah 8y ago|0 comments
131.
by jdwyah 8y ago
FeatureFlags for sure. They don’t force anyone to change their day to day but allow them to adopt better practices at their own speed.
132.
by jdwyah 8y ago
If you have tried R and found it painful I can’t say enough good things about the “R for Data Science” book. Great overview of Tidyverse, ggplot. After 50 pages I was further along than my previous 5 years of googling and cursing.
133.
The last 5 bugs I let slip into production (2009)(blog.jdwyah.com)
2 points by jdwyah 8y ago|0 comments
134.
How to choose a good cache key(blog.prefab.cloud)
1 points by jdwyah 8y ago|0 comments
135.
by jdwyah 8y ago
Also biased, but if you need the functionality and don’t want to run it yourself there is https://www.prefab.cloud/documentation/basic_rate_limits
136.
by jdwyah 8y ago
author here. Definitely legit to consider Heroku and we have used Heroku at some points. In particular since if you build it to work for Heroku it will generally speaking be easy to get to k8s or something else in the future if need be. Ove
137.
How EzCater deploys to kubernetes(engineering.ezcater.com)
7 points by jdwyah 8y ago|2 comments
138.
Comparison of 4 ways you can track an AB Test/experiment(engineering.ezcater.com)
1 points by jdwyah 8y ago|0 comments
139.
Predictive modeling of LTV(engineering.ezcater.com)
1 points by jdwyah 9y ago|0 comments
140.
by jdwyah 9y ago
Zodiac is definitely worth a look. They did a pretty good job of predicting our long-term user’s future revenues. Also all of the academic papers around it are good reads. That said, that didn’t solve our problem because what we really want
141.
Early prediction of LTV for your SaaS(engineering.ezcater.com)
4 points by jdwyah 9y ago|0 comments
142.
by jdwyah 9y ago
One thing that RateLim.it uses to its advantage is calculating the “nextPossiblePass” for each limit. This allows clients to cache forever that something is over the limit until time X and not have to make another request. In the bursty cas
143.
by jdwyah 9y ago
Definitely can negotiate, but it’s not going to be 90%. MixPanel is pretty solidly enterprise. At this level of spend they’re sending a team of a couple implementation people onsite and price is def negotiated.
144.
by jdwyah 9y ago
There's one other "big hammer" for saving money at MixPanel: "send fewer events". We saved 60% (which was no small amount of money) by realizing that sending more than 1 of an event per user per hour was: A) Not ne
145.
Way too much thinking about AB tests and how to track them(engineering.ezcater.com)
1 points by jdwyah 9y ago|0 comments
146.
by jdwyah 9y ago
Chuckle. Same experience. As far as I can tell the estimated time is about as good as the Windows 3.1 file copy time estimation.
147.
Babylonian tablet of base 60 trig(telegraph.co.uk)
3 points by jdwyah 9y ago|0 comments
148.
by jdwyah 9y ago
Worth noting the opportunity cost of not being in S3. Something like Athena won't be yours for the asking. It's been saving my butt lately. Nice to be able to actually see what's in your S3 sometimes :) http://blog
149.
SQL for doing p-value testing in Redshift(engineering.ezcater.com)
2 points by jdwyah 9y ago|0 comments
150.
Log Aggregation at Scale for Cheap with AWS Athena and Kinesis Firehose(blog.ratelim.it)
3 points by jdwyah 9y ago|0 comments

More ›