Demonstrates a good understanding of the business team they are partnering with
E.g. Knows how the team is organised, who is who, what are the goals, current priorities, biggest challenges etc.
Translates business questions into analysable hypothesis and answers those
E.g. Question from business 'Why do salaried users cost us twice as much on customer support?' → cost are allocated by number of intercom queries → salaried users must be generating more queries → Is of queries proportional to engagement? → Are all salaried users are over-proportionally struggling with particular problems (e.g. missing bank statements) → etc.
Picks the right visualisation types for the data at hand
E.g. distributions, time series, scatter plots etc
Basic stats and math knowledge
E.g. Able to find a formula to calculate confidence intervals for different measurement scenarios, knows how to interpret those etc.
Comfortable with using git and contributing to our code base
Can extend existing data models and design simple new ones
Creates new Looker views and dashboards; extracts basic insights quickly from existing Looker explores
Strong SQL skills
Implements basic prediction models quickly
Basic Python or R skills
Delivers assigned tasks that meet expected criteria
Tries to unblock themselves first before seeking help
Works for the team, focuses on tasks that contribute to team goals
Reasons well about about underlying principles of data modeling
Attention to details
E.g. whenever you deliver a piece of work or send a weekly KPIs report you don’t just blindly copy & paste; you sanity check whether things make sense and try to spot mistakes
Manages their own time effectively, prioritises their workload well, on time for meetings, aware when blocking others and unblocks
E.g. able to focus on assigned tasks despite distractions from people, emails, slacks etc. Able to create a 'focus environment' for themselves, exhibits self-awareness around personal productivity (able to spot and debug personal productivity issues or to seek help/advice)
Brings things to completion
Analysts/data scientists often exhibit a behaviour where they run many analyses in parallel for a prolonged time without closing tasks off. Closing a task off could mean writing down key takeaway and sharing the findings with the relevant audience.
Brings a model into a production experiment instead of continuing to tweak offline results.
Data Science: Familiar with ML batch serving techniques
Data Science: Basic knowledge of standard ML approaches
linear regression, neural nets, clustering, random forests etc.
Consistently applies data modeling best practices and suggests ways to improve current practices in non trivial cases
Able determine what really matters for a particular analysis and understands what a 80/20 solution would look like and can prioritise accordingly
Able to pick the best tool and method to effectively help the business to answer a question/make a decision
E.g. Looker, SQL, python or spreadsheets + a basic chart, blackbox ML model or a structured scenario model etc) → Understands the problem at hand and proposes alternative suitable solutions rather trying to fit the problem to the favourite tool.
Concise, clear and effective communication
tailored to audience, clear and concise message (i.e no unnecessary details), can be through emails, slack or presentations
Data Science: Able to pick the right ML method for the problem at hand; demonstrates good intuition of how those approaches work and what strength/weaknesses they have
Data Science: Distinguishes well between impactful ML problems vs just 'predicting something'
Data Analytics: Asks why. Does not take truths for granted unless they understand exactly where they are coming from especially with regards to regulation, compliance, etc.
Actively drives improvements of how the team works
Values teams success over individual success and company’s success over teams success
Onboards / mentors new team members
Gets buy-in on technical decision-making and proposed designs
Sought out for code reviews
Distinguishes clearly between urgent and important tasks and is able to focus on getting the important tasks done.
effectively manages expectations of other people
communicates priorities to their team and other relevant stakeholders
Holds themselves and others accountable
Accountability is about delivering on a commitment. It’s responsibility to an outcome, not just a set of tasks.
Communicates complex ideas effectively
E.g. has the ability to chose the appropriate level of abstraction and make complexity easy to understand tips. more
Data Science: Thrown at fires and resolves / contributes heavily to resolving them
Data Science: Replicates cutting edge approaches from research papers where required
Data Science: Thinks about the future situations code will be used in, planning and acting accordingly
Data Science: Makes pragmatic choices about taking on tech debt
Data Science: Debugs complex Deep Neural Net code/issues.
Knows what to look at when the loss is not decreasing etc.
Data Science: Validates ideas aggressively & iteratively
tackles the biggest unknowns first; validates ideas with 10% effort
Data Science: Measures, understands and is transparent about the impact of their ML work.
we should serve as role models for the rest of the company in this regard in particular
Data Analytics: Valued and trusted business partner for the teams they support
Can be mostly proxied by the type of questions their business partners are asking. 'Can you help me to solve this (hard) problem?' vs 'Can you please pull this number?'
Data Analytics: Proactively identifies relevant/impactful areas for analyses which would deepen the understanding of the business or enable decisions
During the planning process you contribute proactively to help your team to define the right priorities with relevant insights
Solves larger ambiguous/not well defined problems
Contributes to maintaining Monzo’s culture in the wider company
Proactively thinks about how we can get better at our purpose: quicker and better decisions based on data
Builds out a strong internal network
i.e. well connected through-out the company, also to teams with no direct common projects at the moment
Has good organisational awareness
Understands the process of how things are getting done in the company e.g. how and when goals are set, how decisions are being made, how priorities are defined etc.
Sees common patterns in similar tasks and thinks about the solution from the platform/systems perspective.
Solutions that not only solve your own problem but also similar problems of other people in the company)
Data Science: Technical authority within their immediate peer group (team/platform), the natural escalation point
Data Science: Familiar with ML streaming, stateful and stateless serving techniques
can spec out and plan an implementation. Familiar with technological components that might be required
Data Analytics: Deep domain knowledge in specific areas, can go lower than almost anyone else
E.g. deep credit risk knowledge, user behaviour analytics etc
Delivers projects that require cross functional collaboration
Delegates to make better use of their time
Data Science: Serves as a technical authority in the wider data science community
Data Science: Deep domain knowledge, can go lower than almost anyone else
Data Science: Makes targeted improvements in stability, performance and scalability across our platform
Data Science: Measurable impact on company level goals
Data Analytics: Comfortably supports and interacts with C-level executives