Meeting <2017-01-31 Tue>

Learn Better Together

Audrey Troutt

@auditty

≈ 20-70% of time doing a project spent learning

Learning has social component

Chunking info doesn’t matter if you can’t communicate it to someone

Teaching as a component for social learning

Ashley Johnson talk about High Performance Virtual Teams

Solving problems is fun

What most of us got into this industry to do

Learning a New Process

Iteration / Reflection cycle

A means to learn as a group

Learning a New Tool

Hackathons featuring an expert in a framework/language

Learning a New Problem

Pair exchange program with another company

Protected learning time

Company will pay the price of learning one way or another

General learning activity principles

Clarify goals

Apply new knowledge immediately

Protect learning time

Include everyone

Iterate and reflect

bluejeans screensharing

Whiteboard with camera pointed at it

Infrastructure as Code

Jearvon Dharrie

Applying Software Engineering Principles to your Infrastructure

Tools don’t fix the problem

Ansible and Chef seem attractive, but don’t prevent you from using the same practices

Definition

https://en.m.wikipedia.org/wiki/Infrastructure_as_Code

Version control provides audit trail and compliance

Style guide

Infrastructure needs refactoring too

How do you get buy in for refactoring

Martin Fowler suggests do it and don’t tell them

If you touch it leave it better than you found it

Red, green, refactor; even with boxes running NGINX

Always consider cost / benefits with testing

What tool do you use to test Ansible scripts?

Ansible Playbooks

Server spec for Ansible

Nspec for chef

test kitchen for ansible

Outside in testing Gary Bernhardt

https://www.youtube.com/watch?v=tdNnN5yTIeM

Blackbox testing against staging

CLI interface to control publishing in pipeline

Need champion, e.g. to transition from ansible to Docker kubernetes

Meeting <2017-02-28 Tue>

ETE Opportunity Scholarship

A Better Technical Interview

Sam Jones

Test Double

Collective interview

You don’t work in isolation.

Don’t interview in isolation.

Elements of a good interview

Highlight Candidate Experience

Reflect the Job

Give Feedback

If you value communication state it

Don’t just evaluate communication skills without conveying expectations.

Solidify company values

To try to reduce influence of subjective impressions on interviewer.

Reframe No as Not Yet

Create an advocate for the candidate

Clarifies hiring process and values of company

Serves as intermediary among interviewers

Try to cut down on anchoring bias

Learn something from a candidate when pairing

Crafted hiring for diversity

Christopher Litsinger

@x0pherl

Goal to hire the most diverse team possible

Empathy among team makes team good at solving problems

Bias from all-men airbag design teams

Pipeline problem

Greater than half of people coming out of US colleges and universities Computer Science programs are white men.

Most underrepresented population is black females

H1-B visas bring in talent from rest of world

What can industry do to include more people

Hiring may be some of the source of the problem

Interviewing.io

Interview practicing platform

Blog

Can’t measure false negatives in interview process

Methods of interviewing doesnt predict performance on the job

Check the facade you present as if doing opposition research on yourself

Contradictory hiring requirements

Gender disparity in self-selection to apply for these jobs

Apply like a mediocre white guy

Avoid saying things are requirements if they aren’t requirements

Bring 2 of the following 5 skills

s/requirement/preferred qualification/

Be clear and explicit

At the bottom of every job interview:

We are actively seeking a diverse set of candidates to join our team, if you’re not sure if you qualify for this position please err on the side of applying

Define values

Empathy

Ability to learn

Enthusiasm for the job

Raw technical skills

Craft an interview process based on those values

Resumes are a bias quagmire

Added psychological test for empathy

Tests on candidates’ own system; open internet

Pairing in interview process selects for people who have paired before

Blind alley code exercise

Get candidate to have to check ego and consider alternative solution

Wide-band delphi estimation for candidate evaluation

Score against identified values

+2: I will be strong advocate for hiring this candidate

+1: I’m ok if we hire this candidate

-1: I’m ok if we don’t hire this candidate

-2: I will be strong advocate for not hiring this candidate

No neutral value!

People will tend to pick the neutral value. You want their feedback on the candidate they interviewed, so don’t let them pick a neutral value.

Made mistakes

Recruiting through twitter or personal networks

Target for diversity?

Goal: build the most diverse team in Comcast

Can’t really measure outcome

Results of increasing diversity

Some more conversations about accessibility

How do you make an API accessible?