752 karma · joined June 16, 2010
While I personally don't see much use in the particular endpoints you have integrated with, what I do see the need for on a daily basis is API endpoint monitoring.
Modern applications depend on a huge quantity of APIs and this becomes even more true as everything moves to micro services. One endpoint/service going down can break a lot of things as a chain reaction.
There are website monitoring services such as Pingdom but I haven't seen anything dedicated to APIs. You can create a poor man's API monitor with Pingdom (if REST) but once you get inside a corporate network or you need specific authentication, these services don't work anymore.
I can maybe see repurposing what you have built into an API monitoring service which using a dashboards and alerting quickly spots API failure...perhaps?
Each of them held their own code at higher standards than anyone around them and they did it because it was important to them and not to impress others.
At least on Ubuntu if you have a Nvidia GPU there is a handy widget you can install so simply toggle the GPU but it requires a logout/login each and every time.
Just not the same experience as Windows but in terms of battery life it does the job. I get about 3x battery life when in Intel Graphics vs my Nvidia.
Lenovo is completely behind on offering anything remotely pro dev which competes with XPS 15 or MacBook Pro.
I got the XPS 15 9550 for my dad and he loves it.
Is it because the statistics come out in M$ favor when only looking at the 1500 bracket rather than 500 or 2000? Hmm...
Difference from other platforms I've found: integrated machine learning phase which reuses the same data pipelines used by the algo when backtesting or trading live. Trying to reuse as many open source frameworks as possible to reduce the amount of new things a user has to learn. Based on Apache Storm, H2O machine learning, guava, Spring Framework etc. Not HF focused.
Most other platforms start off with you coding up your rules and allow you to back test those assumptions. However, this platform also incorporates the discovery phase of these rules rather than just assume you already have a set of rules.
If it's too much work looking it up yourself, maybe you don't want to code as much as you think you do.
He talked about all the new responsibilities I was going to have an the team members who would report to me, but before he was going to announce it, I just needed to deliver this one thing first. So I did. Then it was another thing. Then he encouraged me to setup 1:1 meeting with my future direct reports. So I did. He however, didn't stop meeting with them directly himself. They were totally confused as to what the heck I was doing and what exactly they were supposed to do with my meetings.
He encouraged me to do sprint planning and work with my future direct reports to estimate stories and execute. Meanwhile, he gave them completely different direction in his own meetings. Soon, no one cared who I was, what I said or whatever was agreed in the sprint planning meetings. That's when he concluded he probably couldn't promote me since I didn't have the teams support and that I probably should just focus on delivery.
Yeah...so that's when I quit.
How do you write/test scripts which are so reliable to work across all the different types of devices, networks, OS etc which participated in this attack? To infect and control this many devices your code must be rock solid.
Over time the reality of the amount of work required kicks in as well as some engineering ego. It's not enough anymore to just get something up and running...it has to be done according to the latest best practices, probably utilizing some new fangled technology I wanted to play with and it needs to be elegant, beautiful and perfect in every way.
The project starts becoming less about accomplishing the original purpose and transforms more into an embodiment of my own engineering ego. Once it shows some early promise, incorporates the latest stack I wanted to play with and either hits a few major design/architecture snags that bog me down or the busy becomes too much...I loose interest.
I hate it and I hate myself for working this way. I am so sick of not finishing what I start and it's the same cycle every time.
Interestingly at work, it's a completely different story. It's all about efficiency, getting the thing done, tested and deployed to production because there are another million things to get done. Super focused, conscious about what is considered good enough and very little patience for other engineers who feel like they need to constantly fiddle with something instead of moving onto bigger more important things.
Just one awesome fact...they structured their code such that they only have one GC event a day (and it's tuned to happen at night). Watch it, pretty cool compared to anyone doing plain vanilla java dev at an enterprise.
Tickerstorm.io - An algo trading platform in Java (pretty early still).
Difference from other platforms I've found: integrated machine learning phase which reuses the same data pipelines used by the algo when backtesting or trading live. Trying to reuse as many open source frameworks as possible to reduce the amount of new things a user has to learn. Based on Apache Storm, H2O machine learning, guava, Spring Framework etc. Not HF focused.
Most other platforms start off with you coding up your rules and allow you to back test those assumptions. However, this platform also incorporates the discovery phase of these rules rather than just assume you already have a set of rules.
If I need to pre-install all these services on my hosts before any of this will work, then how is that really different from uniformly deploying all my applications on an application server such as WebSphere? An EAR too contains all my configuration data, my bindings, my security settings etc. So...if you uniformly run WebSphere everywhere in your enterprise, it sort of accomplishers the same thing.
HOWEVER, if the habitat configuration is capable of talking to my F5 load balancer to create a resource pool, self register my app with my corporate DNS, create a DB schema and populate it with my tables, talk to my networking equipment and open all relevant firewall ports and do so when the application is deployed to a new environment, as well as undo all these actions if it's ever moved elsewhere, then you have piqued my interest...
All the established enterprise companies seem to be eating Docker's lunch by signing huge enterprise deals off of a technology Docker invented and SHOULD have first mover advantage.
If for every position you did:
1. Feature enhancements in HTML, CSS and JS 2. Bug fixes and debugging 3. Wrote scripts, created new layouts etc etc
...then sorry, I am not even interviewing you because you don't seem to be thinking very critically about your job and what you are really here to do for the company.
Always express your resume in terms of accomplishments and what change you produced that affected the ultimate goal of the business.
So no, you didn't do "feature enhancements', you "Increased conversion rate by 5% after acting upon A/B test results. Streamlined the customer on-boarding experience by reducing clutter, pre filling forms, and cutting page latency."
After 8 yeas of experience and a brief listing of technologies you frequently work with somewhere on your resume, I don't really care what technology you used to accomplish the above. The fact that this is how you think about your role at the company is what matters to me. Any deficiencies or strengths will come out during the interview but I'll definitely invite you for a conversation.
And I agree, interviewing front end devs using algorithm puzzles are bullshit. A clear sign the company hiring doesn't know what it actually needs in a person but just do it because it's what Google does (or did).
Do you need batch analysis or stream processing? How much data do you have? How will you need to combine stored data with streaming data? What data needs to be in memory vs disk? How far back in time does you analysis need to go? Can you pull that off in a streaming fashion or do you need a Lambda architecture? Are all your queries known ahead of time or ad-hoc? How will you access historical events? Do you really have data that requires 'big data' tools or can you get away with R on a powerful workstation?
You don't give many details here so I can only through out questions which first need answers.
I.e. in our environment we have the whole Cloudera stack, we use map reduce for some things, spark for others and storm for others. We discovered we don't need Kafka but can get away with just a message broker. We have HDFS, Hive, Couchbase, Oracle and a few other stores all with their own sweet spot.