top of page

Technical Alignment
Research Accelerator (TARA)

A free 14-week course based on ARENA's curriculum, delivering hands-on training in technical AI safety.

Startup Development Team

What is TARA?

TARA is designed to build technical AI safety talent in Australia and New Zealand. Running for 14 weeks (March - May 2025), participants commit to full-day Saturday sessions (9:15 AM - 5:00 PM, Sydney time) plus 2-7 hours of independent learning during the week.

​

You'll learn and implement key concepts like transformer architectures, mechanistic interpretability, and model evaluation techniques. Working in small groups of 4-6 participants, you'll practice pair programming - where two people collaborate on code together, taking turns to write and review - all under the guidance of an experienced Teaching Assistant (TA).
​

The program is completely free—we cover:

  • An expert TA

  • Saturday lunches

  • Compute credits for projects

  • Dedicated study spaces in Sydney and Melbourne (For locations outside Sydney and Melbourne, we'll work with accepted participants to secure suitable Saturday spaces)

Why apply to TARA?

1. You'll work closely with an experienced TA. Through structured Saturday sessions and ongoing Slack support, you'll receive guidance on technical concepts and project development in AI safety.


2. Transform theoretical knowledge into real-world technical skills through guided project work. The program culminates in a project where you'll receive detailed guidance from your TA and peers, with the opportunity to produce publication-quality research.

​

3. We're bringing together 25 participants from across Australia and New Zealand for our inaugural cohort. Through small group sessions, collaborative projects, and ongoing discussions, you'll build lasting connections with others passionate about AI safety. These peer relationships often prove valuable for ongoing work in the field.

Who is this course for?

  • Software engineers and Machine Learning practitioners aiming to transition into AI safety

  • Postgraduate computer science/mathematics students

  • Technical professionals who can’t commit to full-time programs.


These are examples, not requirements. If you have strong technical skills and are passionate about AI safety, we encourage you to apply.​

Locations and dates

TARA runs from March 1 to May 31 2025, with guaranteed groups and Saturday sessions in Sydney and Melbourne. Additional locations across Australia and New Zealand (like Brisbane, Adelaide, Perth, Auckland or Wellington) will be determined by where our strongest applicants are based. 

​

We'll keep applicants updated throughout the selection process. Our goal is to make TARA accessible to talented participants across Australia and New Zealand while maintaining high-quality, in-person study groups.

Curriculum overview (High-level structure)

The program follows a progressive learning journey through technical AI safety concepts:

​

  • Foundations (Weeks 1-3): Build essential technical foundations including optimisation, neural networks, and core ML concepts.

  • Advanced Topics (Weeks 4-6): Dive deep into transformer architectures, mechanistic interpretability, and sparse autoencoders.

  • Applied Learning (Weeks 7-11): Focus on model evaluation techniques and specialized topics, with participants choosing areas for deeper exploration.

  • Project Phase (Weeks 12-14): Apply your learning to develop research-quality projects, with dedicated support from Teaching Assistants.

Weekly learning structure

Each week follows a consistent pattern designed to maximise learning while accommodating participants' schedules:
 

Saturday Sessions (9:15 AM - 5:00 PM):

  • Morning: TA introduces the week's technical concepts and topics

  • Mid-day: Interactive group work and paired programming

  • Afternoon: Office hours, problem-solving, and dedicated TA support

  • End of day: Progress review and planning for independent work

 

Independent Learning (2-7 hours during the week):

  • Self-paced study of concepts introduced on the Saturday

  • Ongoing Slack support from TAs and fellow TARA participants


This structure ensures participants receive structured guidance while maintaining flexibility for individual learning styles and work commitments.

What are the entry requirements?

  • Strong Python programming background

  • Basic understanding of deep learning concepts

  • Working knowledge of linear algebra, probability and statistics

  • Ability to commit to 10 – 15 hours per week for 14 weeks

  • Ability to attend weekly in-person sessions

  • Located in or able to travel to major ANZ cities
     

Even if you don’t fit all these requirements, we encourage you to apply to TARA.

What can I expect from the application process?

You'll spend about 30-40 minutes completing an application form about your technical background, relevant experience, and interest in AI safety. There's no interview process.

​

We use a commitment bond to increase engagement in the program. Here is how it works:

  • You'll put down a $150 AUD deposit when you start.

  • Once you attend your 10th Saturday session (out of 14), you'll receive $165 back (any forfeited bonds will be reinvested into the current TARA cohort).

  • You'll receive your refund within 3 days of reaching your 10th session.

  • If $150 is too much for you right now, we understand. Please let us know in the application form and we can discuss a reduced amount.

Key Dates

  • Participant applications close 23 January 2025.

  • Selected participants notified by 14 February 2025.

  • Week beginning 24 February 2025: participant & TA ice-breaker session (online).

  • Program starts in-person on Saturday 1 March 2025.

  • Black Facebook Icon
bottom of page