Applied AI Engineer

About the product

The product is an AI-native assistant designed to support everyday communication, organization, errands and complex workflows with minimal user input.

It must reliably manage long-running processes, retain context, interact with external tools and complete real-world tasks despite the non-deterministic nature of modern AI models. The objective is to make everyday activities significantly faster and easier for users.

About the role

As an Applied AI Engineer, you will turn model capabilities into reliable product behavior. You will own problems end-to-end—from shaping model behavior and building the surrounding systems to ensuring that everything performs effectively in production.

The position sits at the intersection of Machine Learning, systems engineering and product development. The focus is on making AI deliver real value to users in production environments—not only in prototypes and demonstrations.

What you will do

• Build and ship AI features end-to-end, from the model and supporting systems to the user experience.
• Design and continuously improve prompts, tools, memory systems and agent workflows.
• Transform raw model outputs into structured, reliable and predictable product behavior.
• Diagnose and resolve issues across models, orchestration, infrastructure and user experience.
• Optimize systems for latency, operational cost and production reliability.
• Develop lightweight evaluation frameworks to measure real-world model and product performance.
• Work closely with product and engineering teams to transform ambiguous problems into production-ready systems.

Technology stack

• Python
• PyTorch / JAX
• LLMs, including OpenAI-compatible APIs, LLaMA and Qwen
• Model inference and serving, including vLLM
• Vector databases

What we are looking for

• Strong foundations in Machine Learning and modern neural network architectures.
• Hands-on experience training, fine-tuning or deploying Machine Learning models.
• Ability to write clean, maintainable and production-quality code.
• Confidence working across multiple abstraction layers—from models and infrastructure to product behavior.
• Strong problem-solving skills in ambiguous and rapidly changing environments.
• A delivery-oriented mindset focused on shipping, measuring, iterating and continuously improving production systems.

What success looks like

• Production Machine Learning models consistently meet accuracy, latency and reliability expectations.
• Production issues are detected quickly, diagnosed effectively and resolved at the root-cause level.
• Data pipelines, training workflows and inference systems are robust, reproducible and maintainable.
• ML-powered functionality is delivered through effective collaboration with engineering, product and research teams.
• Improvements to models and systems are driven by real-world signals, measurable outcomes and user feedback.

Compensation and employment

The position is offered under an employment contract.

Compensation is determined individually based on:

• Professional experience and technical capability
• Scope of responsibility
• Location and relevant market benchmarks
• Expected impact on the product and organization

The compensation package consists of a base salary and equity. Flexibility is available for exceptional candidates.

A company laptop will be provided where required for the role.

Remote work and global collaboration

The organization operates remotely with globally distributed teams. There is no fixed company-wide working schedule or requirement to follow one specific time zone; team members are expected to maintain sufficient working-hours overlap with immediate colleagues to collaborate effectively.

The successful candidate will work from Poland and collaborate with Machine Learning, engineering, product and research specialists located across different regions. Poland-based employees join existing global teams rather than a separate local team. The exact reporting line and hiring manager will be confirmed during the recruitment process.

How the team works

The team is small, highly capable and hands-on. Decisions are made collaboratively, while individuals are expected to bring structure to ambiguous problems, exercise sound judgment and execute independently.

The team moves quickly while balancing production quality, experimentation and continuous learning from real-world usage.

There is no fixed hiring quota for this position. Engineers are selected based on technical and ownership standards rather than a predetermined number of openings.

Recruitment process

The standard recruitment process consists of up to four stages:

• Technical assessment, where relevant to the candidate’s background
• HR interview
• One or more technical interviews
• Founder or leadership interview

Particularly strong candidates may be fast-tracked directly to the technical interview stage based on their experience and previous work.

Applications are evaluated by members of the technical team. Interviews may be conducted virtually and, where relevant, onsite. The exact format and interviewers may vary.

The aim is to make decisions efficiently and provide candidates with a prompt outcome.