About the job:
Performance & Scale: Design, build, and optimize AI-driven systems that process complex natural language interactions. Delivering low-latency, high-quality responses from our AI Assistant requires high-availability and smart optimization of LLM and retrieval workloads in practice.
Modern Tech Stack: Work with a world-class, cloud-native polyglot environment. Our core backend technologies include Java, Kotlin, Spring, and ADK. At the same time, our AI and orchestration layer heavily relies on Python. You will utilize Python and LangChain for complex LLM orchestration, alongside Langfuse for observability, prompt tracking, and evaluation. We leverage advanced vector search and modern ML solutions to make our assistant truly smart.
True Ownership & Autonomy: We live by a "you build it, you run it" philosophy. You'll join an agile, autonomous team with full ownership of your AI microservices and data pipelines, from design to deployment; and improvements are dictated by results of our own experiments and AI evaluation metrics.
Complex Challenges: From LLM orchestration and Retrieval-Augmented Generation (RAG) to semantic search optimization, you'll tackle the complex backend problems that power a conversational marketplace assistant. You will work across the Java/Kotlin and Python ecosystems, integrating advanced embedding models, prompt pipelines, and vector databases to ensure our AI deeply understands and accurately retrieves information.
In our domain, we focus on delivering the core search and retrieval capabilities for the Allegro AI Assistant. We maintain a critical entry point for users looking to discover products, compare offers, and get personalized recommendations through natural conversations. Striving for relevancy and accuracy first, we are designing, implementing, and operating an extensive system combining modern microservice architecture with Large Language Models, high-performance retrievers, and vast data sources. We care about data quality, robust context building, and near real-time accuracy to best guide users through a highly dynamic marketplace via an intelligent, conversational interface.
Important things for you:
Flexible working hours in the hybrid model (4/1) - working hours start between 7:00 a.m. and 10:00 a.m. We also have 30 days of occasional remote work.
The salary range for this position depending on the skill set is as follows (contract of employment, tax-deductible cost):
PLN 14 200 - 20 200
Annual bonus based on your annual performance and company results.
Our team is based in Poznań.
We are looking for people who:
Code in Python, Java and Kotlin;
Expect work that requires creative approach to search experience challenges, to be proud of enjoying its results together with all Allegro users;
Want to create solid code, eagerly accept comments and share knowledge during code reviews;
Can look for effective solutions to the requirements set by users;
Want to constantly develop and update their knowledge;
Know English at at least B2 level.
Nice to have: You are open to diverse technologies and workflows, including dev-operational tools on one hand, and working with data on the other. Basic experience with LLM tuning** **(LoRA, PEFT), training and evaluation pipelines for models and/or basic frameworks for building AI Agents (Agent Development Kit, LangChain, AutoGen, CrewAI) would be great!
What's in it for you:
Well-located offices (with e.g. fully equipped kitchens, bicycle parking, terraces full of greenery) and excellent work tools (e.g., raised desks, ergonomic chairs, interactive conference rooms).
A 16" or 14" MacBook Pro or corresponding Dell with Windows (if you don't like Macs) and all the necessary accessories.
A wide selection of fringe benefits in a cafeteria plan - you choose what you like (e.g., medical, sports or lunch packages, insurance, purchase vouchers).
English classes that we pay for related to the specific nature of your job.
A training budget, inter-team tourism ( see more here ), hackathons, and an internal learning platform where you will find multiple trainings.
An additional day off for volunteering, which you can use alone, with a team, or with a larger group of people connected by a common goal.
Social events for Allegro people - Spin Kilometers, Family Day, Fat Thursday, Advent of Code, and many other occasions we enjoy.
And that's just the beginning! You can read more about the benefits here .
#goodtobehere means that:
**You will join a team you can count on **- we work with top-class specialists who have knowledge- and experience-sharing in their DNA.
You will love our level of autonomy in team organization, the space for continuous development, and the opportunity to try new things. You get to choose which technology solves the problem and you are responsible for what you create.
You will value our Developer Experience and the full platform of tools and technologies that make creating software easier. We rely on an internal ecosystem based on self-service and widely used tools such as Kubernetes, Docker, Consul, GitHub, and GitHub Actions. Thanks to this, you can contribute to Allegro from your very first days on the job.
You will be equipped with modern AI tools to automate repetitive tasks, allowing you to focus on developing new services and refining existing ones (also leveraging AI support).
You will create solutions that will be used (and loved!) by your friends, family and millions of our customers.
You will meet the Allegro Scale, which starts with over 1000 microservices, an open-source data bus (Hermes) with 300K+ rps, a Service Mesh with 1M+ rps, tens of petabytes of data, and production-used machine learning.
You will become part of Allegro Tech - We speak at industry conferences, cooperate with tech communities, run our own blog (it's been over 10 years!), record podcasts, lead guilds, and we organize our own internal conference - the Allegro Tech Meeting. We create solutions we love (and can) to talk about!
Send us your CV and… see you at Allegro!