Junior+ ML Engineer (Model Architecture, Training and Edge Optimisation)
First seen by Alion on Oct 16, 2026.
Full-time
Salary: $1000—$1,500/month, reviewed after 3 months based on delivery
Position is for recent university graduates majored in Machine Learning, Mathematics or related field.
Prior work experience is not required, but deep understanding of machine learning and mathematics under the bonnet is.
About CubicMind AI
CubicMind AI Ltd is a UK deep tech startup building a small plug-in device that operates any computer like a person: it reads the screen, listens to the user, and clicks and types to complete tasks. Everything runs on the device itself, with no cloud.
cubicmind-ai.com
We design our own models rather than wrapping someone else’s. That includes transformer and non-transformer architectures, built around what runs fast on edge hardware, and sometimes trained from scratch.
About the role
You will work on designing, training and optimising the models that power the device: screen understanding, action prediction and efficient on-chip inference.
This is research-heavy engineering. You will work on model architecture, training runs, attention and memory bottlenecks, low-bit compression, and getting models to run correctly on real hardware.
What you will do
— Design and implement new model architectures in PyTorch, both transformer-based and beyond
— Train models, sometimes from scratch, and run experiments to compare architectures
— Build training and evaluation pipelines on screen-understanding and computer-control data
— Profile latency, memory use and KV-cache behaviour on long, high-resolution screen inputs
— Compress models for edge chips using quantisation (INT8/INT4), pruning and distillation
— Export models to low-level runtimes and fix unsupported operators
— Work with our C++ team to get models running efficiently on the device
Requirements
— Architectural understanding of Transformers and Attention: for example you understand the mathematical and operational differences between MHA, MQA, GQA, FlashAttention, and linear attention variants;
— Low-level PyTorch fluency: writing custom PyTorch modules, vectorised tensor operations, e.g. knowing how to do ML
— Experience with model training & convergence: stability tricks, learning rate schedules, gradient clipping, mixed-precision training (BF16/FP16), and diagnosing loss spikes or exploding/vanishing gradients.
— Mathematical foundation: comfortable with high-dimensional linear algebra, matrix calculus, multivariable optimization, and probability theory.
— Understanding what model compression is: Post-Training Quantization vs Quantization-Aware Training, outlier activations, etc
— Сommit full-time
Nice to have
— Experience with vision transformers or vision-language models
— Interest in non-transformer architectures (state-space models, linear attention, recurrent designs)
— Experience exporting to ONNX, TensorRT, OpenVINO or NPU toolchains
— Basic C++, enough to read inference code
What we offer
— Real ownership of the core models in a hardware product built from the ground up
— Work directly with the founder, with no bureaucracy
— $1000—$1,500/month depending on experience, reviewed after 3 months based on results
— Fully remote, judged on results, not hours logged. Flexible schedule