AI Systems Engineer

Future Beyond

Salary: Gross salary $6400 - 8000
Type: Full time

Tags: Microservices Machine Learning API Development Containers

Beware: we're not looking for a "Prompt Engineer". We're looking for an AI Systems Engineer: someone who maintains a clear mind that thinks in terms of architectures, latency, and scale, knowing how to pull tricks out of thin air to make AI work in the messiest real-world conditions. Always searching for the sweet spot that lives between simple and complex, between under and over-engineering.

Exclusive offer from getonbrd.com.

What you'll actually be doing

You'll be part of the core team crafting Future Beyond’s building blocks. Your daily battles won't be tuning hyperparameters in an isolated notebook. They'll be:

Orchestration. Assembling all the pieces into a coherent, complex system. You'll make architecture decisions with vendor-agnosticism — picking the right technology for each subsystem without falling for hype. That means designing execution graphs (LangGraph, custom DAGs, or whatever the problem demands) where LLMs reason alongside classical ML models that detect, classify, optimize, recommend, and forecast. You'll decide when a problem needs a fine-tuned foundation model and when a GBM or a time-series model is the sharper tool — then wire them together with resilience, robustness, and security built in from the start.
Traceability. Building systems where every AI recommendation is fully auditable. In the industries we work in, a black box is unacceptable. Our software must explain its inputs and reasoning chain transparently — whether that chain passes through an LLM, an anomaly-detection model, or both.
Physical-digital integration. Connecting AI and ML models with a diversity of data streams — IoT telemetry, spatial databases, other models' outputs — to build systems that suggest best-next-actions in real time. You'll train and deploy classical inference pipelines (predictive maintenance, sensor-fusion classifiers, geospatial clustering) that feed into and are orchestrated by higher-level AI agents.
From prototype to production. Taking AI proofs of concept and turning them into resilient, fault-tolerant modules that run 24/7 in the world's most demanding environments. That includes owning the full lifecycle of ML models — feature engineering, training, validation, deployment, monitoring for drift.

What we expect from you

You're an engineer who builds production-ready AI systems. You have solid backend experience and a deep understanding of software architecture. You are language-agnostic but expert in high-performance stacks, comfortable with strict typing, asynchronous programming, and building resilient APIs that handle mission-critical workloads.
Infrastructure is not a mystery to you. You understand how the cloud is actually built. You have real fluency with containers (Docker, K8s), databases (SQL, no-SQL, and vector search), and microservices architectures.
You write and read the source code. You have an obsession to understand how things work underneath. You strongly use AI as a co-pilot to go faster, never as autopilot. You own every line of code you put into production.
Deliberate friction. Against the current wave of technological noise, we hold a deliberate contradiction: we're naive enough to explore, and cynical enough to commit. We want someone who's quick to get their hands dirty with a new model, and ruthlessly skeptical before adding it to the stack.
You communicate as well as you code. Because we're autonomous and "Humans First," being able to effectively and respectfully communicate with others is a must. You can write a tech spec, a GitHub issue, or a Slack message with total clarity — in both English and Spanish. The same applies to explaining technical decisions to a non-technical stakeholder, or pushing back on an architecture choice in a design review.

Bonus points

(none of these are required, but they'll make us lean forward)

Experience building AI systems with machine learning models, optimization, recommendation, autonomous agents or advanced LLM architectures (complex RAG, LangChain/LangGraph, LLM evaluation).
You've done something — or remember doing something — with low-level languages (Go, Rust, C++), or have a working understanding of high-performance computing and memory management.
Experience with time-series data, industrial telemetry (IoT), or mathematical modelling.
Familiarity with ML deployment methodologies in production (MLOps, CI/CD for ML).

Benefits

4-day week. We work Monday through Thursday, 8 hours a day. We believe in the balance between work and life — and nothing reinforces that like a 3-day weekend every week. We work 80% of the hours and maintain 100% of the salary and output (We suspect that we're actually more productive by working one day less).
32-hour work week. Some weeks require more, but those are the exception, not the rhythm.
20 days of vacation per year.
Work from wherever you want. You're always welcome at our offices in Valdivia and Santiago.
Extended parental leave — 6 months full salary for mothers, 1 month for fathers.
Apple equipment.
Learning budget to get certified and keep sharpening your skills.
Access to any tool you need to do your work better.
Remote workspace budget to set up your home setup properly.
Health and dental insurance.
Sport and mental health stipend.
Profit-sharing program based on tenure, starting after 2 years.

Fully remote You can work from anywhere in the world.

Pet-friendly Pets are welcome at the premises.

Flexible hours Flexible schedule and freedom for attending family needs or personal errands.

Health coverage Future Beyond pays or copays health insurance for employees.

Computer provided Future Beyond provides a computer for your work.

Informal dress code No dress code is enforced.

Vacation over legal Future Beyond gives you paid vacations over the legal minimum.

Beverages and snacks Future Beyond offers beverages and snacks for free consumption.

APPLY VIA WEB

Source: GetOnBoard | Main Category: Other