The Intersection of Product and Engineering Teams

TLDR: Product and Engineering working together closely is a super power.

Thoughts about working from home/remote

As you know, the COVID-19 virus is underway and more people are starting to work from home. I love going into the office, but ***I never really stopped to understand WHY I love it so much, and if the...

Testing Alert Service Failover

Today I spent a chunk of time testing the message queue failover system, which is part of Amazon MQ's high-availability (HA) features. Every message broker with this feature enabled has an active & s...

For our testing, we rebooted the MQ broker and made sure both our message producers and consumers actively switched between the available broker urls we give them. It took us about 4 hours of reconne...

Database Studies #1

I've decided to do a deeper study into SQL style databases because currently it's a hole in my knowledge, and I feel it will super useful in the future. First I need to organize my studies & find inf...

If an information source and/or tutorial isn't "working" for you personally, throw it away immediately and find another.

Small Update

I only have time for a small update today. Been pretty swamped with meetings and performance enhancements for the dashboard. Feeling very confident in the way both the team and the product is going,...

Here's a SS what we've been working on:

Setting Up Sprints

As a tech lead, I've been bequeathed with the role as "technical project manager", which probably has various manifestations at different companies & at different times. In our case right now, I've d...

The meeting went well! We made some minor adjustments to my sprint vision, and I noted them down. Then I added the new team sprint standard into our knowledge base solution notion.

Message Queue Failovers

*** AmazonMQ provides an "active/standby broker" system for High Availability (HA) to act as redundant pairs spread between availability zones in case of local catastrophic events such as local floods...

Today was a lot of catchup after a long release process and getting the stomach flu and being out for 2 days. One of the catchup items was implementing the MQ failover features into the publisher and...

Release Day #2

Dashboard and API successfully released! Very happy and tired, so won't be going into many details. It was a lot of work, and the team all worked together very well.

Some issues came up in the last few days before the release , especially around the display of more than 100 of an alert types on the dashboard. We hadn't fully designed and implemented a "load more" ...

Release Day #1

Holy cow, yesterday turned out to be so so busy! It was part one of our release process for our new Inflight feature set, which includes the alert service and queue system I've been talking about for...

As a full stack developer, I'm constantly hopping back and forth between things like messages queues and html, which are very VERY different types of beasts. Interestingly enough, the majority of my t...

Article Writing

One of my tasks today is to start a technical/narrative article around our move to a message queue. First thing to do is nail down the potential audience and find some things that would provide value...

Dockerizing Alert Consumer & Adding to AWS

Today, our devops (Developer Operations) guy and I got the alert consumer service running in staging -

First we 'dockerized' & setup configuration on the python project. This involves writing a dockerfile in the root of the project that extends our base image for python projects, and adding some glue ...

Alert Service Update

Today was hectic, but productive. What got done:

1. Got some important device alert code into the staging API. This involved creating a MR (merge request) for my new code, getting it reviewed by a colleague, making any fixes/updates, rebasing, and...

Wrapping up the week

Curation of the Asana project board is the name of the game this morning. Created the last remaining tasks for our new Alert/Eureka features. It looks like this:

Found a clever solution to something that I wanted to humbly brag about real quick

I have this grid I made in Sketch with gradient dividers. Something tricky to actually build in code:

Diving into the user API

User API work today - the alert service will hit the user API for 3 operations in the course of it's daily business:

1. Getting SIM/link information. (Exists)

We be body movin... err, I mean data gathering

The last couple days, I've been brainstorming how the alert service will gather data it needs to make the decision to create an alert for a device. **Goal is to get the most up to date usage for the...

Test Entry

Today I made my first entry in my Hologram work journal.

Alert Service Redis Setup

This day marks the start of our final sprint (2 weeks) before we do a big feature release on Jan. 30th. The UI & API portion of the work is coming along very nicely, but the python alert consumer is ...

Heads down on Message Queue & Alert Service

Working from home today. We all informally announce our WFH intentions in a Slack #general channel, which has been a good system so far.

The rest of my day will be “heads down” on the new alerts service, which is the impetus for the better stronger message bus system talked about previously. The alert service itself is in python, and ...

Message Queue Show & Tell

Just arrived at work. I usually have a checklist when i first arrive, which is “Email, Slack, Asana, Journal”. So i’ve checked my email & slack, and did some asana ticket management (moved a few tick...

First, I’ll be leading a “show and tell” meeting at 3pm with most of the engineering team at hologram. Primary goal is to give everyone, even down to the junior developers, a clearer understanding of...

It's been a while...

Hi all,

It’s good to be back writing another dev blog! It’s been a couple years since my last one, but it’s time to do it again.