Backend Engineer, AI Agent Systems
⚲ Warszawa, Śródmieście
16 000 – 22 000 zł / mies.
Wymagania
- Python
- Node.js
- PyTorch
Opis stanowiska
Nasze wymagania:
Strong backend engineering fundamentals supported by experience with production systems.
Experience building or operating high-throughput, low-latency services.
Familiarity with AI inference patterns, including LLMs, embeddings and multimodal models.
Confidence debugging distributed systems under load.
Understanding of production observability, incident management and performance optimization.
A delivery-oriented mindset focused on shipping, observing production behavior and improving through iteration.
Ability to work independently and make pragmatic engineering decisions in an ambiguous environment.
O projekcie:
About the product:
Our client is developing an AI-native assistant designed to support everyday communication, organization, errands and complex workflows with minimal user input.
The product 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 company’s objective is to make everyday activities significantly faster and easier for users.
About the role:
As a Backend Engineer specializing in AI systems, you will own the inference and orchestration layer behind the product’s AI interactions.
Your work will sit between models and end users, where latency, correctness, reliability and operating costs directly affect the product experience.
You will build and operate production systems that transform model capabilities into fast, stable and observable APIs used by mobile and desktop applications.
What success looks like?
• Backend systems reliably handle production AI traffic at scale with low latency and high throughput.
• APIs remain stable, clear and easy to integrate with frontend, mobile and Machine Learning systems.
• Production incidents are detected quickly, diagnosed effectively and resolved with minimal impact on users.
• System performance and reliability improve continuously based on real-world usage and production data.
Technology stack:
• Python
• Node.js
• PyTorch
• OpenAI, Anthropic and open-source LLMs
• SQL and NoSQL databases
• Kubernetes
• Docker
Zakres obowiązków:
Build and operate backend systems serving AI-powered functionality in production.
Design inference pipelines, orchestration layers and service boundaries around Machine Learning models.
Take ownership of production monitoring, logging, alerting and incident response.
Optimize latency and throughput across inference, caching, batching and streaming.
Build reliable APIs supporting integration with mobile, frontend and Machine Learning systems.
Diagnose distributed-system issues under production load.
Continuously improve system performance and reliability using real-world production signals.
Strong backend engineering fundamentals supported by experience with production systems.
Experience building or operating high-throughput, low-latency services.
Familiarity with AI inference patterns, including LLMs, embeddings and multimodal models.
Confidence debugging distributed systems under load.
Understanding of production observability, incident management and performance optimization.
A delivery-oriented mindset focused on shipping, observing production behavior and improving through iteration.
Ability to work independently and make pragmatic engineering decisions in an ambiguous environment.
O projekcie:
About the product:
Our client is developing an AI-native assistant designed to support everyday communication, organization, errands and complex workflows with minimal user input.
The product 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 company’s objective is to make everyday activities significantly faster and easier for users.
About the role:
As a Backend Engineer specializing in AI systems, you will own the inference and orchestration layer behind the product’s AI interactions.
Your work will sit between models and end users, where latency, correctness, reliability and operating costs directly affect the product experience.
You will build and operate production systems that transform model capabilities into fast, stable and observable APIs used by mobile and desktop applications.
What success looks like?
• Backend systems reliably handle production AI traffic at scale with low latency and high throughput.
• APIs remain stable, clear and easy to integrate with frontend, mobile and Machine Learning systems.
• Production incidents are detected quickly, diagnosed effectively and resolved with minimal impact on users.
• System performance and reliability improve continuously based on real-world usage and production data.
Technology stack:
• Python
• Node.js
• PyTorch
• OpenAI, Anthropic and open-source LLMs
• SQL and NoSQL databases
• Kubernetes
• Docker
Zakres obowiązków:
Build and operate backend systems serving AI-powered functionality in production.
Design inference pipelines, orchestration layers and service boundaries around Machine Learning models.
Take ownership of production monitoring, logging, alerting and incident response.
Optimize latency and throughput across inference, caching, batching and streaming.
Build reliable APIs supporting integration with mobile, frontend and Machine Learning systems.
Diagnose distributed-system issues under production load.
Continuously improve system performance and reliability using real-world production signals.
🔍 Dekoder Ogłoszenia
🟡
delivery-oriented mindset focused on shipping
Nacisk na szybkie wdrażanie kodu, możliwe pomijanie pełnej dokumentacji/testów
🔴
work independently and make pragmatic engineering decisions in an ambiguous environment
Mało wsparcia ze strony zespołu, niejasne wymagania, trzeba samemu radzić sobie z niepewnością
🟡
high-throughput, low-latency services
Wymagania wydajnościowe będą rygorystyczne, praca pod presją na optymalizację
🔴
debugging distributed systems under load
Częste awarie produkcyjne i praca pod presją w sytuacjach kryzysowych