Software Engineer (Foundational AI)

Brak informacji o wynagrodzeniu

MidZdalnie·ManisesFull-time·Umowa o pracę

#471062·Dodano 5 dni temu·2

Źródło: justjoin.it

Aplikuj teraz

Tech Stack / Keywords

PyTorch depthPythonvLLMSGLangquantisation

Podsumowanie oferty

Join Maisa’s AI team to train, fine-tune, and evaluate models that power Maisa Studio. The work focuses on Python and PyTorch, supporting deployment for clients in regulated industries.

Wymagania

Current, hands-on experience training and fine-tuning models.
Strong PyTorch skills, including building and modifying model architectures directly.
A rigorous approach to model evaluation, including detecting regressions and degradation.
Solid software engineering fundamentals in Python.
Sound judgment when weighing custom models against accuracy, cost, and latency.
Comfort working in a fast-moving, ambiguous environment with a high degree of ownership.
Strong communication skills for working with engineers, product, and leadership.

Nice to have:

Experience with model serving and inference optimisation at scale using vLLM, SGLang, or quantisation.
Experience deploying models in regulated, on-premise, or air-gapped environments.
Familiarity with the Hugging Face ecosystem, including transformers, datasets, PEFT, and TRL.
Experience with distributed or multi-GPU training.
Open-source contributions or published work in the ML community.
Background in an early-stage or scaling startup.

Obowiązki

Train, fine-tune, and post-train Maisa's small and large language models, from data preparation through evaluation and release.
Design and run model evaluations, including benchmarks, regression sets, and quality criteria based on client use cases.
Build and improve training pipelines and infrastructure, including data curation, experiment tracking, and reproducibility.
Apply and adapt training techniques such as parameter-efficient fine-tuning, post-training, and alignment methods, and evaluate their impact.
Work with the AI infrastructure function on model serving and inference optimisation using vLLM and SGLang, including deployment constraints inside client environments.
Monitor deployed models for quality degradation and drift, and drive retraining or corrective work.
Help shape model strategy, including where custom models add value compared with frontier models or deterministic code.
Collaborate with Applied AI, product, and engineering teams to ensure models meet product and client needs.

Maisa AI

6 aktywnych ofert

Zobacz wszystkie oferty

Aplikuj teraz