AI Engineer
⚲ Warszawa, Mokotów
26 880–35 280 zł netto (+ VAT) / mies.
Wymagania
- Python
- LangGraph
- LlamaIndex
- CrewAI
Opis stanowiska
Nasze wymagania:
Min. 5 years of experience as an Engineer in the IT industry (preferably as a Backend/Python Engineer)
Being able to work until at least 7:00 PM CET.
Experience building applications on top of LLMs (OpenAI, Anthropic, or similar)
Hands-on commercial experience developing AI Agents
Strong Python skills and familiarity with agent frameworks (e.g. LangGraph, LlamaIndex, CrewAI)
Understanding of prompt engineering, structured outputs, and context management
Experience with vector databases and retrieval systems
Familiarity with evaluation and testing approaches for non-deterministic systems
Ability to reason about tradeoffs between prompting, RAG, and fine-tuning
Mile widziane:
A degree in Computer Science, Artificial Intelligence, or a related field
O projekcie:
Hi there! If you’re looking for a high-impact position in an ambitious software house, we’ve got a match for you!
Currently, we are searching for a full-time AI Engineer to cooperate with our US client - the company integrates production-grade, governed AI workflows into complex enterprise systems by addressing common friction points like poor integration and lack of domain expertise. The organization combines deep domain knowledge with cutting-edge capabilities to accelerate the deployment of reliable AI products.
Our client’s core mission is to make artificial intelligence work for complex enterprise workflows. Recognizing that most AI initiatives fail due to poor integration, gaps in domain specialization, a lack of governance, and unclear ownership-rather than flawed models—they focus on overcoming these exact friction points.
By combining decades of deep industry domain expertise with cutting-edge product AI capabilities, they accelerate production deployment and embed production-grade, governed AI workflows directly into some of the world’s most intricate enterprise systems. The leadership team consists of proven industry veterans with a track record of delivering groundbreaking AI products, backed by a board of senior executives with deep expertise in global consulting and software engineering.
Due to the client's location, a daily standup around 5:00 – 6:00 PM Polish time, with required availability to work until at least 7:00 PM CET
Zakres obowiązków:
Design and implement the agent orchestration architecture, defining how agents are structured into tiers, how they pass work to one another, and how cross-domain workflows are decomposed and coordinated.
Build the domain harness the scaffolding layer that defines each agent's input and output contracts, grounding context, approval enforcement hooks, and error recovery logic.
Implement tool integration patterns using the MCP protocol, including tool schema definitions, call routing, response handling, and fallback behavior when tools are unavailable
Define and implement Human-in-the-Loop (HITL) checkpoint
Build the context injection mechanism that retrieves relevant policy documents, tenant configuration, and conversational memory and injects them into agent prompts before each LLM call
Oferujemy:
Private medical care
Multisport card
Friendly, informal atmosphere, and direct contact with everyone in the company
Modern technologies
Min. 5 years of experience as an Engineer in the IT industry (preferably as a Backend/Python Engineer)
Being able to work until at least 7:00 PM CET.
Experience building applications on top of LLMs (OpenAI, Anthropic, or similar)
Hands-on commercial experience developing AI Agents
Strong Python skills and familiarity with agent frameworks (e.g. LangGraph, LlamaIndex, CrewAI)
Understanding of prompt engineering, structured outputs, and context management
Experience with vector databases and retrieval systems
Familiarity with evaluation and testing approaches for non-deterministic systems
Ability to reason about tradeoffs between prompting, RAG, and fine-tuning
Mile widziane:
A degree in Computer Science, Artificial Intelligence, or a related field
O projekcie:
Hi there! If you’re looking for a high-impact position in an ambitious software house, we’ve got a match for you!
Currently, we are searching for a full-time AI Engineer to cooperate with our US client - the company integrates production-grade, governed AI workflows into complex enterprise systems by addressing common friction points like poor integration and lack of domain expertise. The organization combines deep domain knowledge with cutting-edge capabilities to accelerate the deployment of reliable AI products.
Our client’s core mission is to make artificial intelligence work for complex enterprise workflows. Recognizing that most AI initiatives fail due to poor integration, gaps in domain specialization, a lack of governance, and unclear ownership-rather than flawed models—they focus on overcoming these exact friction points.
By combining decades of deep industry domain expertise with cutting-edge product AI capabilities, they accelerate production deployment and embed production-grade, governed AI workflows directly into some of the world’s most intricate enterprise systems. The leadership team consists of proven industry veterans with a track record of delivering groundbreaking AI products, backed by a board of senior executives with deep expertise in global consulting and software engineering.
Due to the client's location, a daily standup around 5:00 – 6:00 PM Polish time, with required availability to work until at least 7:00 PM CET
Zakres obowiązków:
Design and implement the agent orchestration architecture, defining how agents are structured into tiers, how they pass work to one another, and how cross-domain workflows are decomposed and coordinated.
Build the domain harness the scaffolding layer that defines each agent's input and output contracts, grounding context, approval enforcement hooks, and error recovery logic.
Implement tool integration patterns using the MCP protocol, including tool schema definitions, call routing, response handling, and fallback behavior when tools are unavailable
Define and implement Human-in-the-Loop (HITL) checkpoint
Build the context injection mechanism that retrieves relevant policy documents, tenant configuration, and conversational memory and injects them into agent prompts before each LLM call
Oferujemy:
Private medical care
Multisport card
Friendly, informal atmosphere, and direct contact with everyone in the company
Modern technologies
🔍 Dekoder Ogłoszenia
🔴
Being able to work until at least 7:00 PM CET.
Oczekiwane są regularne nadgodziny, które mogą wykraczać poza standardowe godziny pracy.
🔴
high-impact position
Może oznaczać dużą odpowiedzialność i presję na wyniki, a niekoniecznie realny wpływ na kierunek projektu.
🔴
ambitious software house
Może sugerować szybkie tempo pracy, częste zmiany priorytetów i potencjalnie brak stabilności.
🟡
production-grade, governed AI workflows
Wymaga to nie tylko umiejętności technicznych, ale także zrozumienia procesów biznesowych i regulacyjnych wdrożeń AI.
🟡
addressing common friction points like poor integration and lack of domain expertise
Praca może polegać na rozwiązywaniu problemów, które już istnieją w systemach klienta, a nie na tworzeniu czegoś od podstaw.