Titulo Estágio
Large Language Models Inference
Áreas de especialidade
Engenharia de Software
Sistemas Inteligentes
Local do Estágio
Coimbra
Enquadramento
LLMs can bring value to Feedzai customers in a variety of ways, also within a variety of solutions. This brings the need for our Machine Learning infrastructure to accommodate such models. The challenge we propose aims to 1) enable Feedzai to effortlessly and agnostically interface with external LLM providers and 2) serve Feedzai-owned LLM in a fully-fledged way.
Objetivo
LLMs can bring value to Feedzai customers in a variety of ways, also within a variety of solutions. This brings the need for our Machine Learning infrastructure to accommodate such models. The challenge we propose aims to 1) enable Feedzai to effortlessly and agnostically interface with external LLM providers and 2) serve Feedzai-owned LLM in a fully-fledged way.
Plano de Trabalhos - Semestre 1
"0. Onboarding on Feedzai's Model Serving
1. Discovery of existing generic LLM APIs (Feedzai has its own, MLFlow has a generic one as well, etc.)
2. Discovery of self-hosting frameworks
3. Assessment of privacy, limitations, prices and performance (latency & throughput) in self-hosting vs remotely hosted solutions
Expected output:
- Literature review document
- Plan approach for second semester based on those findings"
Plano de Trabalhos - Semestre 2
"Expected results:
- Implement self-hosted Proof-of-Concept deployed on kubernetes within Feedzai's ML Infra ecosystem
- Assess its price and performance in throughput and latency
- Documentation, final report and presentation"
Condições
Remunerated
Orientador
Alberto Ferreira
alberto.ferreira@feedzai.com 📩