# Objectives for 01/07/23 - 31/05/24
- Martin Stoffel

## Objective 1

**Objective**: Create an MVP for the emulator platform [`autoemulate`](https://github.com/alan-turing-institute/autoemulate).

**Category**: Reinforcing the role of the Institute as a national body in the UK and International landscape

**Target**:
Background: The idea behind `autoemulate` is to make it easy for engineers and researchers to replace compute-intensive simulations in Digital Twins (and beyond) with fast, efficient emulator models. To achieve this, the package fits and evaluates a wide range of machine learning models to find the best emulator model, which can then be used for prediction, uncertainty quantification and sensitivity analysis.

I'm currently the main developer of `autoemulate`, but I have regular meetings with Eric Daub and Steve Niederer and I'm collaborating with Research Application Managers and other researchers within the DT-TRIC to make the package as useful as possible.

The goal here is to create an MVP, which can be downloaded and tested, which will consist of:

- at least five different state-of-the-art machine learning (emulator) models
- automatic cross-validation
- automatic hyperparameter search,
- visual and tabular model diagnostics
- parallelisation.
- substantial unit testing (code coverage > 80%).

**Target date**: mid March, 2024

## Objective 2

**Objective**: Improve `autoemulate`'s impact as open source software.

**Category**: Reinforcing the role of the Institute as a national body in the UK and International landscape

**Target**:
We aim for `autoemulate` to have impact beyond the existing project through collaboration, good documentation and outreach. To reach that goal, I will:

- ensure that `autoemulate` has a clear contributing guide for the community
- create easy-to-follow documentation for users, including demos and examples
- give least one talk at a seminar or conference
- have a RAM assessement of the project with Sophie and Kalle with the goal to improve impact

**Target date**: mid April, 2024

## Objective 3

**Objective**: Set good EDI standards for `autoemulate`

**Category**: Fostering inclusivity and diversity within REG and the Turing

**Target**:
I'll research and best practices around EDI in open source software development, which I will incorporate in the contributing guidelines for the package. To do this, I will collaborate/get feedback from RAMs and/or colleagues in the EDI service area.

**Target date**: mid April, 2024

## Objective 4

**Objective**: Project management and leadership for the`autoemulate` project.

**Category**: Reinforcing the role of the Institute as a national body in the UK and International landscape

**Target**:
As the main developer on `autoemulate`, I'll use the opportunity to develop my project management and leadership skills through:

- creating and maintaining a GitHub project board (in collaboration with Kalle Westerling)
- breaking down the project into specific tasks and milestones
- setting up meetings with different groups in the DT-TRIC and beyond to:
* understand their emulation use cases and where `autoemulate` can help
* collect relevant datasets / simulators to benchmark `autoemulate`
- closely collaborate with the PI and REG lead to determine the project direction and decide what to include / not to include in the MVP
- mentor enrichment student (Bryan Li), who is keen to work on the deep learning side of the project

**Target date**: mid April, 2024

## Objective 5

**Objective**: Create data safe haven course for DSG participants.

**Category**: Development, delivery and support of the Turing 2.0 strategic priorities

**Target**:
Within the Applied Skills Service Area, we (Jules, Ryan, David SJ and I) are developing a moodle course for Data Study Group (DSG) participants to learn about Data Safe Havens (DSH). The course will lead the participants through the basics of DSH's and explain how to use the Turing trusted research environment. The goal is to create an MVP of the course until December so that the participants in the December DSG can be the first cohort testing and giving feedback on the course.

**Target date**: January, 2023

## Objective 6

**Objective**: Faciltate open source practices and development and the Turing

**Category**: Building a culture of "Team Turing" and demonstration of our stated values

**Target**:
After the DSH course is finished, I will transition to the Open Source Service Area where I will:

* help to organise monthly hack sessions, where Turing colleagues are contributing to upstream open source repos. These hack sessions will hopefully contribute to both the personal development of Turing employees by working on different projects, but also to improve the wider impact of the Turing institute through contributions to the wider open source ecosystem.
* start to create a tool for analysing GitHub data. The goal of the tool is to get a better understanding of how the Turing contributes to the wider open source ecosystem, but also to learn more about the various open source projects within the Turing.

**Target date**: May, 2023

# PDP

### What skills do I have? What are my current strengths?

* quantitative research mindset: quickly getting up to speed with new projects by being curious, reading, trying things out using code / visualisation and asking questions
* statistics
* data science programming: data wrangling, exploration, visualisation, modeling
* collaboration: listening to people and asking questions, figuring out common goals, discussions
* big picture thinking: critically reflecting on projects and thinking about the wider impact and vision

### What skills do I want to get? What are my area of further development?

* software engineering: best practices for large codebases
* deep learning: getting more practice in writing and deploying deep learning models
* leadership: getting more experience in managing and mentoring people
* project management: learning the fundamentals and specific techniques, such as scrum

### How am I using my current skills and strengths?

My current projects leaves me plenty of room to use my current skills. I'm working on an open ended project spanning simulations from various disciplines, a wide range of machine learning models and software engineering. I'm also collaborating with different researchers.

### What opportunities are there to develop my skillset and strengths? How would I use them?

- Projectwork: In my project, I'm touching upon many areas where I'd like to improve my knowledge / skillset. For areas where this is high priority, I can just take some more time to learn about them properly or set up discussions with colleagues who know more about it.
- Reading groups: there are a variety of reading groups at the Turing to learn about topics like foundation models, linear algebra or bitcoin. I'm joining those groups when I've got time.
- Courses: I'd be interested in project management, leadership and deep learning courses, if I find useful courses and have the budget and time. I'm going to explore some of the possibilities here.

### What could potentially get in the way of my Professional Development?

I'm currently the only developer on my project, so I'm feeling some time pressure to get things done. Therefore, a focus on driving the project forward could get in the way of working on my wider professional development.

# Professional Development Goals

### What do I want to focus my Professional Development on this year?

I'd like to focus on developing four areas:

* Project management (scrum, project boards)
* Software engineering (best practices for large python codebases)
* Deep learning (getting better at PyTorch)
* Leadership

### What support and resources will I need?

* Opportunities for leadership
* Ressources to develop project management
* Budget to go to conferences / courses
* Time

### How will I measure success?

I'll measure success in two ways:
* having taking courses to improve in these areas
* having successfully applied new skills in different parts of my Turing projects

### Target dates for review and completion