Agentic Edge AI Engineer

5V Tech; Eindhoven; On-site; EUR 70,000 - 110,000; Full time

Benefits offered by this job

Competitive Salary

Reward scheme for referrals

Job summary

A leading semiconductor company in Eindhoven is looking for an AI Engineer specializing in the optimization of LLMs and edge AI applications. The role involves hands-on engineering with advanced models and optimizing them for reliable on-device deployment. Candidates should have strong skills in C/C++, Python, and a solid understanding of Agentic AI systems. This position offers a mid-senior level opportunity with a competitive salary and full-time employment.

Qualifications

• Strong hands-on engineering skills in C/C++, Python, Bash on Linux systems.
• Proven technical background in LLM/LMM/VLMs, system optimization, and Agentic AI.
• Experience with model optimization and performance tuning for edge/embedded/NPU targets.

Responsibilities

• Optimize LLMs/LMMs/VLA models for on-device deployment.
• Accelerate inference with system-level optimizations.
• Engineer safety-compliant agents with guardrails and secure workflow design.
• Deploy optimized models to achieve efficient NPU-accelerated inference.
• Build benchmarking pipelines for performance metrics.
• Deliver PoCs for edge use cases.

Skills

C/C++

Python

Bash

System optimization

Agentic AI

AI toolchains

Tools

llama.cpp

Ollama

ONNX Runtime

TFLite

Job description

Agentic Edge AI Engineer

• Initial 12-months contract + option to join as an employee
• Competitive Salary / Rate depending on experience level

A leading semiconductor company is expanding a high-priority R&D program focused on Agentic and Generative AI at the edge. This team is taking cutting‑edge research and turning it into production‑ready, on‑device capabilities.

They are searching for an AI Engineer to work hands‑on with LLMs, large multimodal models (LMMs), and Vision‑Language‑Action (VLA) / Audio‑Language‑Action systems, ensuring they run reliably and efficiently on NPU‑based edge platforms using advanced optimization techniques.

A key part of the role is building agentic workflows that are safe by design: function calling, tool orchestration, policy enforcement, guardrails, and secure tool invocation patterns that can operate under embedded constraints.

What you’ll do...

• Optimize LLMs/LMMs/VLA models for on‑device deployment via advanced quantization, pruning, distillation, and bandwidth/memory reduction.
• Accelerate inference with system‑level optimizations (e.g., speculative decoding, efficient decoding strategies, KV‑cache tuning) built for resource‑constrained edge environments.
• Engineer “tiny” safety‑compliant agents: tool use and function calling with guardrails, safe invocation, and secure workflow design.
• Deploy optimized models via llama.cpp, Ollama, ONNX Runtime, and TFLite to achieve efficient NPU‑accelerated inference.
• Build benchmarking pipelines for latency, memory footprint, energy efficiency, and reliability on real target hardware.
• Deliver PoCs for edge use cases (e.g., industrial safety monitoring, in‑cabin sensing).

What you’ll bring

• Strong hands‑on engineering skills in C/C++, Python, Bash on Linux systems.
• Proven technical background in LLM/LMM/VLMs, system optimization, and Agentic AI.
• Strong knowledge of AI toolchains, deployment, and inference engines.
• Familiarity with embedded system development and tooling.
• Proven experience with model optimization and performance tuning for edge/embedded/NPU targets.

This is the perfect opportunity for someone with experience in both LLMs and Edge AI, who is looking to work on truly cutting‑edge development.

APPLY NOW to discuss the details further...

5V Tech are acting as an Employment Agency for the purposes of this position. We offer a reward scheme if you can recommend someone for this position; €250 for you and €250 to a charity of your choice.

Seniority level

• Mid‑Senior level

Employment type

• Full‑time

Job function

• Engineering and Information Technology
• Semiconductor Manufacturing