On April 5th, I attended StaffPlus 2022 New York conference. This was my fifth time attending an event organized by LeadDev. I’ve previously attended four of their conferences in London, including the first one ever. I’m happy to see they’ve continued to grow as a community and an event organizer.

I’m in New York for a different conference organized by LeadDev happening the same week - LeadingEng. I’ve visited StaffPlus because while it isn’t directly aimed at me, I thought I’ll still be able to receive a lot of value. And I did.

On-demand access to all talks is available from April 12th, 2022. In the meanwhile, I share my notes from the talks. I don’t necessarily agree with everything I share here, but I found all of it exciting and compelling. Hopefully, they’ll give you a sense of what StaffPlus is all about. My notes are listed below in chronological order.

Silvia Botros, Being a principal engineer. The world is your oyster.

• Traditional move for a long while was to go from senior engineer to a manager
  • Managers assumed that those who can do could also teach
  • Engineers assumed more decision-making power
  • People chose to go into management for the wrong reasons
• Enter “Staff+ Career ladders” for senior individual contributors
  • Technical leader: makes technical decisions and leads
  • Don’t manage people, but participate with input in evaluating teams’ performance
  • Translate product strategy to technical strategy and technical execution
  • Say “it depends” a lot, as there’s a lot of ambiguity
• Enterprise Principal Engineering Basics:
  • Just because you can doesn’t mean you should.
  • How can I help?
  • What are the consequences of coupling these systems?
  • How will we interface with X if Y goes down?
  • Is this written down anywhere?
• Beware the stereotypes and actively work against them
  • “The Ivory Tower”, “Bring the purple robes”
  • They exist from real scar tissue, don’t dismiss them
  • You run the risk of losing touch with how the team works: be with the team and be involved

Ryan Harter , Strategies for focusing on the Most Important Thing

• Bad defaults: prioritization by inertia, decibels or guilt
• The ritual: empty your inbox, reflect on the last week, choose your work
• The questions that help with the ritual:
  • What can you delegate?
  • How can you delegate more?
  • What should you be working on?
  • What “must” you do?
  • What feels urgent? Is It?
  • Who cares? What happens if I drop this?
  • How would you convince the CFO they can’t afford to lose you?
  • What do you take home at the end of the week?
  • What would change everything?

Amy Unger, Starting a new job as a Staff+

• Seven key questions that help figure it out
• 1: Why were you hired?
  • Why was this position created at this level, and not a level below?
  • Who are you looking to fill this position externally?
  • “Nobody wanted this job”
• 2: How can I build my network?
  • Build organizational context, walk through the structure
  • Do 1:1s with ICs who report into your manager
  • Build team context and your peer network
• 3: What’s your job?
  • Will Larson’s archetypes: tech lead, architect, solver, right hand
  • Discuss how you’re spending your time
  • Clarifying your manager’s job can help you understand yours
  • What kind of work gets people promoted?
• 4: What (wrong) lessons did you learn?
• 5: What do you tackle first?
  • Work backward from your goals
  • Your manager’s job will always be easier if you are more confident
• 6: What can you uniquely do as a new person?
  • As a new person, you have opportunities you won’t have later
  • If it’s unclear to you as a new joiner, it’s likely unclear to someone more junior
• 7: When do you know enough?

Michelle Bu, Lifecycle of an API design

• Write the quickstart guide: akin to writing the press release
• Model the domain: the heart of software is its ability to solve domain-related problems for its user
• Package the model
• Test and iterate: build a lot of integrations, identify friction, and remove friction

Katie Sylor-Miller, Moon to Mars: Planning and executing for the long term

• The Architect’s dilemma: how do you plan, execute, and get buy-in for large-scope projects?
• Example story: the PWA (Progressive Web Apps) Project
  • PWAs drive significant business results
  • Another dilemma: rebuild or refactor?
  • “We’re gonna be slower to market!” is not a great sell
• To get to Mars, first, you have to land on the Moon
  • Partner with teams outside of your org
  • Infrastructure investment enables future work: break down work so that each phase enables the next one and provides value on its own
  • Get feedback early and often: continuous measurement and testing
  • Bold vision: architect as a force multiplier
• Skills to level up your team
  • Navigate ambiguity inherent in long timelines
  • Understand when and why to change plans
  • Think about tradeoffs in both the short- and long-term
  • Use data to drive decisions

Mattie Toia, Getting to Commitment: Tackling broad technical problems in large organizations

• Know your goal: the first person you convince is yourself
• Focus on values: tap into the deeper meaning
• Address your audience: in a way that meets their needs
• Present the progress: it’s an opportunity to define success
• Open your mind: to the ideas of others

J. Bobby Dorius, Solving inefficiency challenges at Datacenter scale

• As an SRE, I don’t want to be PAGED because we didn’t plan for capacity
• Twitter runs their own data center, including building their hardware
• Tens of thousands of computers were like regular cars with HUGE engines: how do we get two vehicles to use a single engine?
• Dividing up compute and storage workload at the socket level
• 9-figures in CAPEX savings: small innovative change with tens of thousands of servers

Leslie Chapman, Becoming an influencer (no selfies required)

• How to influence without having direct reports and budgets?
• Name recognition is one way to recognize influencers: Kim Kardashian, Paris Hilton
• Bragging is a way to create name recognition: tell people what you’ve done
• Personal brand: give people different ways to remember you
• Building trust by getting things done leads to people thinking “I can’t wait to work with them”
• Learn to say “no” otherwise you won’t be able to get the rights things done
• Listen, aggregate, give credit and create something greater than the individual parts
• We need to be the people to help the next generation.

Alice Li, Accessibility and why everything old is new again

• ADA Ramps as example for “solve for one, extend to many”
• Semantinc HTML elements are important not only for developers, but for readers as well
• The lang HTML attribute helps the screenreaders figure out how to pronounce the text
• Dark Mode is the native environment for screens: energy-efficient, ergonomic, accessible
• Respect user preferences: the right thing to do might not be the most popular one

Scott Triglia, Driving clarity in ambiguous projects

• Ambiguous projects can be ambiguous for different reasons: unclear requirements, complex plans, complicated team dynamics
• Before the project begins
  • Resist poorly defined goals with a PRFAQ (product brief)
  • Be explicit about your customers and their specific needs
• Getting off the ground
  • All stakeholders are not built alike, be specific about the stakeholders (e.g., RACI)
  • Start building from a tight nucleus, keep the team small at the start
  • Warning sign: it takes months to finalize the perfect design
  • Spur discovery with mediocre proposals, but be explicit
• Turning the corner on execution
  • Build, measure, learn: execute against your plan, adjust when you learn new information
  • Avoid the lake
  • Decisions should be first-class, logged, and broadcasted in public
  • Consider using documents, not email
• Getting it done
  • Pick the finish line of your project scope, and repeat it a lot
  • Getting projects across the finish line is a challenge
• Managing yourself
  • If the project leader is panicking, everyone else will start panicking too
  • Allowing yourself to bend and adjust with pressure means you won’t break

John Riviello, The art & science of decision buy-in

• How to decide, and who will decide?
  • There are many ways to decide: democracy, meritocracy, autocracy
  • Multiple independent judgments > expert’s individual judgment
  • Disagreement: one or both parties are likely to be wrong
• Conclusions from the Harward Business Review study:
  • Pre-commit to a strategy for combining their opinions
  • The person responsible for final judgment shouldn’t form an opinion beforehand
  • Disagreement should be thought of as valuable information
• Analytic hierarchy process (AHP): goal, criteria, alternatives
  • Worked-through example showing the use of AHP
  • Discussions might be more important than the numbers themselves
  • AHP Tool created by John
• Nemawashi: a semi-formal consensus-building procedure
  • Make the decisions slowly, execute them quickly

Bryan Liles, Building a better bridge

• Pro tip: alliteration is awesome
• When instead of a road, a swamp is all around you: congrats, you’ve become a staff engineer
• Boss and leader might be the same, or it might not be the same
• How to be a 10x engineer: find five engineers and make them twice as productive
• 7 levels of delegation
• We can only control the inputs, but not the outputs
• Simon Sinek: The Golden Circle

David Daly, Leveraging your peers as a Staff Plus Engineer: Building peer groups

• Keep it lightweight when starting to build a community of peers
• Get the scope right: both too few and too many people won’t work
• Self-organize: the members of the group should run the group
• Provide value: the group should be serving the purposes of the members