LLM Engineer

Job description

Company Description

Welcome to the good side of tech 👋

You might have heard about us, but with a different name: Znanylekarz. It all started 10 years ago when we asked ourselves: is anyone in healthcare thinking about patients? We jumped in and we empowered patients by giving them access to leave and read reviews about their visit. We then provided doctors with the technology to manage bookings easily and save time, so they could devote themselves to what they always wanted: treating patients. And today is the day in which we ask you: wanna join us in the next step of making the healthcare experience more human?

Docplanner at scale

We are leaders in 13 countries so far, and more than 90 million patients trust us every month. 300.000+ specialists believe in us and our product, and so do leading venture capital funds such as Point Nine Capital, Goldman Sachs Asset Management and One Peak Partners. And yet, employing over 2.500 people all over the globe, we managed to keep the startup-mindset we started with over 10 years ago.

How does Docplanner Tech fit here?

At Docplanner Tech we are a diverse group of over 400 people working in Engineering, Data, and Product teams. We are responsible for building the product for all locations. Many of us have been here for over 5 years, yet we still welcome each new person with great joy and excitement.

We could tell you about us, but we will let our reviews on Glassdoor speak for themselves. In case you’d like to see how it feels to be 100% yourself at work, here’s a video of us.

And why should you join us?

Because it feels good to tell your family and your friends how you made the world a little bit better. You go to bed knowing that what you do matters, and that your talents align with your beliefs.

We want to make the healthcare experience more human, and that starts with you being you. We believe that taking the diversity of human experience into account makes a better healthcare experience for all . We’re not just different: we embrace diversity. We will encourage you to come to work your whole self, and that includes not coming to the office at all if you prefer not to, as we’re 100% remote-friendly.

Job Description

Area of Work

You will join our global Machine Learning and Data Science unit — a core team of machine learning scientists (LLM, ASR), engineers and an expanding group of country-specific linguists. Your work will be embedded in the ever-growing Noa product line used daily by doctors in Brazil, Mexico, Poland, Italy, Spain and Germany. Noa.ai’s mission is to broaden access to top-quality healthcare, first by lifting administrative burdens from clinicians and soon, by assuming selected medical tasks. Our flagship product is Noa Notes, which records consultations, transcribes the audio and produces a structured draft of the medical record.

We operate cross-functionally, moving ideas from prototypes to full-scale production with state-of-the-art ML frameworks and modern MLOps practices. We begin with rapid validation sprints where engineers work side by side with Product on early experiments, then consolidate and scale proven capabilities, standardizing validation methods, engineering best practices, tools and lifecycles.

Role

As an LLM Engineer in the Noa ML team you will support a product area within our Noa organisation to deliver end-to-end LLM capabilities. You will work alongside other machine learning professionals at various seniority levels and report directly to the Head of Machine Learning & Data Science.

Backed by a strong engineering culture, we pair industry-leading LLM implementation rigor with pragmatic delivery, making smart trade-offs to ship value quickly and iterate fast. In this role, you’ll work with exceptional engineers and other LLM researchers on some of Docplanner’s most strategic initiatives, leveraging foundation models to push the boundaries of AI in healthcare.

Our tech stack includes Python, LangChain, HuggingFace Transformers, Vector Databases, FastAPI — among the others — running on Kubernetes in AWS.

What you will be doing

• Work closely with cross-functional teams, including scientists, engineers, and product stakeholders, to deliver LLM-driven initiatives that directly contribute to business objectives.

• Design, deploy and iterate over LLM services for text-based applications (and beyond), while proactively identifying and eliminating performance bottlenecks.

• Build small to medium-sized Python projects and collaborate with engineers on production code and deployments at scale.

• Assess platform engineering and LLMOps bottlenecks; research and design scalable prompt management strategies, and recommend solutions that balance performance, cost, and reliability.

• Research, architect, and deploy LLM-powered information retrieval solutions (e.g., RAG) to deliver accurate results in complex, multilingual product environments.

• Partner with the AI Platform team to refine LLMOps best practices, evolve frameworks, and establish efficient, scalable workflows.

Qualifications

You’re likely a great fit for this role if you

• At least one year of professional experience in LLM development or integration in a fast-paced, product-driven tech environment.

• Demonstrated expertise in production-grade LLM deployments, including prompt management systems, vector databases, semantic search implementation, and API integration with foundation models.

• Good understanding of transformer architectures and proficiency in LLM frameworks such as LangChain, LlamaIndex, or similar tools.

• Proficiency in Python

• Experience in collaborative project development.

• Appreciation for good engineering practices and maintainable code

• Proven experience in evaluating LLMs through systematic testing, benchmark design, and the development of custom metrics (e.g. accuracy, consistency, factuality, and bias), with a focus on aligning results to product and user needs.

• Proven ability to integrate, deploy, and optimize large language models in production-grade industry environments, ensuring scalability and robust performance.

• Strong knowledge in prompt engineering, agent-based workflows, and the generation and manipulation of embeddings.

• Experience with RAG (Retrieval-Augmented Generation) techniques, vector similarity search, and information retrieval methods to enhance LLM capabilities.

• Problem-solving mindset and adaptability in dynamic environments with a focus on delivering business value to end customers.

• Proven ability to manage timelines, prioritize tasks, and deliver results under tight deadlines.

• Curiosity and eagerness to collaborate with cross-functional team

Additional Information

Compensation structure

• A salary adequate for your experience and skills.
• Share options plan after 6 months of working with us.

True flexibility and work-life balance

• Remote or hybrid work model with or hub in Warsaw;
• Flexible working hours (fully flexible, as in most cases you only have to be on a couple of meetings weekly);
• ²⁰⁄₂₆ days of paid time off (depending on your contract);
• Additional paid day off on your birthday or work anniversary (you choose what you want to celebrate).

Health comes first

• Private healthcare plan with Signal Iduna for you and subsidized for your family.
• Multisport card co-financing for you to have access to sports facilities across Poland.
• Access to iFeel, a technological platform for mental wellness offering online psychological support and counseling.

Keep growing with us

• Free English classes.

We promote and embrace equal opportunities in our hiring process, and also every day at work. When you apply for our roles you receive equal treatment regardless of age, disabilities, gender reassignment, marital or civil partner status, pregnancy or parental status, race, color, nationality, ethnic or national origin, religion or belief, sex, sexual orientation or any other dimension of human difference. If you require additional support in your recruitment process, we kindly encourage you to let us know. Behind those words you’re reading, there’s a person (hi!) who already helped a candidate by adapting the interviews, and now we’re lucky to have this person with us. So, even if you’ve never asked for it before, may this serve as a sign that, now, you can do so. We can only truly be equal if we adapt to each other.

“We believe all humans, in all their beautiful diversity, should have equal rights, dignity and respect. Period.” Mariusz Gralewski, CEO