AI Software Development Intern

68.9k PLN68 880 PLN/ rok./ rok.StażUmowa o staż

InternHybryda·GdańskPart-time·Umowa o staż

#463074·Dodano 4 dni temu·6

Źródło: Intel

Aplikuj teraz

Tech Stack / Keywords

vLLMSGLangOpenVINOPython Linux AWS Google Cloud Platform Jenkins PostgreSQL Kubernetes

Podsumowanie oferty

As part of Intel’s Datacenter and AI Software Ecosystem Engineering team, you will evaluate, benchmark, and optimize Generative AI and LLM inference workloads across Intel platforms. You will develop and validate AI serving solutions using vLLM, SGLang, and OpenVINO, with work spanning performance analysis and cloud-native deployment.

Wymagania

Strong background in artificial intelligence, machine learning, computer science, or a related field.
Proficiency in Python and experience developing software applications and automation tools.
Understanding of transformer-based generative AI models, inference workflows, tokenization, quantization, and serving architectures.
Experience working in Linux environments, including shell scripting, system monitoring, and troubleshooting.
Currently enrolled in a Bachelor’s, Master’s, or PhD program in computer science, artificial intelligence, software engineering, or a related field.

Nice to have:

Experience with an LLM serving framework such as vLLM, SGLang, TensorRT-LLM, TGI, or similar.
Experience benchmarking and performance-tuning AI inference workloads.
Experience with AWS and/or Google Cloud Platform services.
Understanding of CI/CD concepts and tools such as Jenkins, Git, and containerized workflows.
Knowledge of model optimization techniques, including quantization, pruning, KV-cache optimization, and speculative decoding.
Experience with container orchestration technologies such as Kubernetes and Docker.
Experience building APIs and services using Python frameworks such as FastAPI.
Personal or academic projects involving LLM deployment, retrieval-augmented generation (RAG), AI agents, or inference optimization.

Obowiązki

Deploy and evaluate large language models using serving frameworks such as vLLM, SGLang, and OpenVINO.
Measure and optimize serving metrics, including throughput, latency, TTFT, TPOT, concurrency, and resource utilization.
Develop automated benchmarking workflows to characterize model performance across hardware and software configurations.
Deploy and evaluate AI workloads in AWS and Google Cloud Platform environments.
Build CI/CD pipelines, test automation, and infrastructure tooling using Jenkins, Python, and Linux.
Design and maintain systems for storing, querying, and analyzing benchmarking and telemetry data using PostgreSQL and related technologies.
Investigate trade-offs between performance, scalability, accuracy, and cost across deployment scenarios.

Inne informacje

The role follows Shift 1 (Poland).

Intel

22 aktywne oferty

Zobacz wszystkie oferty

Aplikuj teraz