About Frontal

Frontal is building the infrastructure for enterprise intelligence.

We’re creating an AI-native platform that enables enterprises to turn their data, systems, and operations into intelligent software that can understand, reason, decide, and act. Our stack spans AI infrastructure, models, agents, data, evaluation, observability, and the runtime required to operate intelligent systems reliably in production.

We’re a small, engineering-driven team based in Lisbon, working on problems that sit at the intersection of machine learning, distributed systems, and enterprise software.

About the role

As an AI/ML Engineer, you’ll work across the full machine learning stack, including model experimentation, training, inference, evaluation, and production deployment.

You’ll help build the intelligence layer behind Frontal: models, agent systems, evaluation infrastructure, retrieval and reasoning systems, and the runtime that makes them reliable at enterprise scale.

This is not a role where you’ll simply integrate third-party APIs. You’ll be expected to understand how modern AI systems work, experiment rapidly, and turn research and prototypes into production-grade infrastructure.

You’ll work closely with the founding team and engineers across the platform to determine what we build, how we build it, and where we can push the technology further.

What you will do

Design, build, and deploy production machine learning and AI systems.
Develop and improve LLM-powered systems, including reasoning, retrieval, tool use, agents, and structured generation.
Work with open and proprietary models across training, fine-tuning, inference, and evaluation.
Build model evaluation and benchmarking infrastructure to measure quality, reliability, latency, and cost.
Develop inference pipelines and optimize models for production workloads.
Experiment with fine-tuning, distillation, quantization, embeddings, reranking, and other model optimization techniques.
Build data pipelines and datasets for training, evaluation, and experimentation.
Develop systems for model observability, tracing, monitoring, and failure analysis.
Work on agent architectures, including orchestration, memory, planning, tool execution, and multi-agent systems.
Improve the performance and reliability of AI workloads running across GPUs, containers, and distributed infrastructure.
Translate research papers and emerging techniques into working systems.
Design and run experiments, analyze results, and make decisions based on empirical evidence.
Contribute to the architecture of Frontal’s AI platform and underlying runtime.
Write production-quality code and the infrastructure required to operate it reliably.
Collaborate closely with platform, product, and infrastructure engineers.

About you

You can reason clearly about machine learning, deep learning, and the systems that make them useful.
You are highly proficient in Python and comfortable working in a production engineering environment.
You have hands-on experience with modern deep learning frameworks such as PyTorch, JAX, or equivalent.
You understand transformers, attention mechanisms, embeddings, tokenization, and modern generative AI architectures.
You understand how ML systems behave beyond the notebook, including inference, latency, memory, scaling, and reliability.
You are comfortable reading research papers and turning ideas into experiments and implementations.
You care about measurement and know how to design meaningful evaluations rather than relying on intuition.
You can move between research-oriented experimentation and production engineering.
You are comfortable working independently in an early-stage environment where the requirements are not always defined for you.
You have strong software engineering fundamentals: Git, testing, debugging, APIs, systems design, and code quality.
You care about building simple, robust systems rather than unnecessary abstractions.
You are curious about how things work at a fundamental level and naturally go deep when solving difficult problems.
You want significant ownership over what you build.

Bonus if you

Have worked with LLMs, multimodal models, diffusion models, or other foundation models.
Have trained, fine-tuned, or served models yourself.
Have experience with CUDA, GPU programming, or inference optimization.
Have worked with distributed training or distributed inference.
Have experience with vLLM, TensorRT-LLM, SGLang, Ray, DeepSpeed, or similar systems.
Have built agentic systems or complex tool-use workflows.
Have experience building evaluation frameworks, benchmarks, or datasets.
Have contributed to open-source ML or systems projects.
Have published research, participated in ML competitions, or built significant ML projects independently.
Have experience with Kubernetes, containers, AWS, or other cloud infrastructure.
Have strong Rust, C++, or systems programming experience.
Follow the latest developments in AI research and regularly experiment with new models and techniques.

Benefits

Build foundational AI infrastructure at an early-stage company.
High ownership and direct influence over the product and technical direction.
Work alongside a small, highly technical engineering team in Lisbon.
Competitive salary and meaningful equity.
Access to modern GPU and cloud infrastructure for experimentation and production.
Work on real enterprise AI workloads rather than toy applications.
Rapid learning and exposure across machine learning, infrastructure, agents, and distributed systems.
A small, high-context team that values direct communication, fast feedback, and building.
Company support for conferences, technical events, and open-source contributions.
The opportunity to help shape the engineering culture and systems of Frontal from the ground up.

How we hire

We hire slowly and keep the bar high. An open seat is better than a compromise, and every person who joins should make Frontal better. We look for people who learn unfamiliar territory quickly, take ownership of outcomes, and can move from an idea to a shipped result.

We value builders over narrow executors: people who understand why the work matters, exercise product judgment, and make the whole company stronger.

Our process is practical and work-based. We will ask you to reason through an unfamiliar technical problem, make tradeoffs, and build or critique something. Curiosity and evidence of craft matter more than university names, titles, or a polished résumé. A résumé is optional, never required; send a short note about the work you are proudest of and links that help us understand how you think.

We expect high standards from one another without creating a culture of bureaucracy. Strong people should be able to operate with autonomy, say “I don’t know,” challenge assumptions, change their minds, and remain honest about what is true. Integrity and clear communication matter as much as technical ability.

Frontal should be a place where exceptional people become substantially better. We want to build the company that ambitious builders in Portugal most want to spend the important years of their careers contributing to. That starts with the quality of the work, people, and products.