Software Engineer – Agentic AI

Impala Search Ltd; Berlin; On-site; EUR 70,000 - 110,000; Full time

Job summary

Impala Search Ltd in Berlin is seeking a Software Engineer – Agentic AI to help build production-grade multi-agent systems, backend infrastructure and data architectures powering AI-driven professional services.

You will own end-to-end delivery, shape architectures, collaborate with product stakeholders, and translate business needs into robust software while advancing agent coordination, memory, retrieval, and orchestration.

Qualifications

• Solid backend engineering fundamentals with production experience.
• Hands-on experience building AI agents or agentic AI components (orchestration, context handling, memory, retrieval).
• Proficiency in Python, TypeScript or equivalent, with experience in backend frameworks.
• Familiarity with agent development frameworks (LangGraph, LangChain, CrewAI or similar).
• Experience building backend services with FastAPI, NestJS, Spring Boot or similar.
• Working knowledge of PostgreSQL, Docker, Git and GitHub; plus cloud/container environments (Kubernetes, AWS).
• Strong German communication skills at C1 level or native.
• Ownership mindset and ability to turn loosely defined requirements into working software.

Responsibilities

• Build the core product by developing an internal platform to streamline complex business workflows.
• Engineer intelligent agents, focusing on coordination, context management, memory, and retrieval.
• Shape the architecture with emphasis on application design, orchestration, APIs, data structures, security, and technology choices.
• Develop reliable data infrastructure, designing and improving relational and vector-based storage for efficient AI data use.
• Own delivery end-to-end by translating business needs into technical solutions and fostering maintainable practices.
• Collaborate with product stakeholders and users to prioritise improvements and deliver value in short development cycles.

Skills

Backend development fundamentals

AI agent experience

Python/TypeScript

Agent frameworks (LangGraph LangChain)

Backend frameworks (FastAPI NestJS S-B

PostgreSQL Docker Git

Cloud platforms (Kubernetes AWS)

German language (C1)

Ownership mindset

Tools

LangGraph

LangChain

CrewAI

FastAPI

NestJS

Spring Boot

PostgreSQL

Docker

Git

GitHub

Kubernetes

AWS

Job description

Our client is reimagining how professional services are delivered by combining human expertise with intelligent AI technology. With 450+ businesses served across 20+ industries, they are now building proprietary AI systems designed to automate complex workflows, empower specialists and turn traditional service delivery into scalable software.

Role Overview

You’ll join as a Software Engineer – Agentic AI, building production-grade multi-agent systems, backend infrastructure and data architectures that power the next generation of AI-driven professional services.

Key Responsibilities

• Build the core product: Develop and enhance an internal platform that streamlines complex business workflows, turning user feedback and operational challenges into practical software features.
• Engineer intelligent agents: Extend production AI systems by working on agent coordination, context management, memory, retrieval and iterative improvement mechanisms.
• Shape the architecture: Influence technical decisions across application design, workflow orchestration, APIs, data structures, security and technology selection.
• Develop reliable data infrastructure: Design and improve relational and vector-based data storage, ensuring information can be retrieved efficiently and used effectively by AI applications.
• Own delivery end to end: Translate business needs into technical solutions, take responsibility for implementation and help establish maintainable engineering practices as the platform evolves.
• Collaborate across disciplines: Work closely with product stakeholders and internal users to prioritise improvements, validate solutions and deliver value through short development cycles.

Qualifications

• A solid professional background in software engineering, with strong backend development fundamentals and experience delivering production applications.
• Practical experience building AI agents or agentic AI components, particularly involving orchestration, context handling, memory or retrieval.
• Proficiency in Python, TypeScript or a comparable programming language, together with experience using established backend frameworks.
• Familiarity with agent development frameworks such as LangGraph, LangChain, CrewAI or comparable tools.
• Experience building backend services with technologies such as FastAPI, NestJS, Spring Boot or similar frameworks.
• Working knowledge of PostgreSQL, Docker, Git and GitHub.
• Experience with cloud-based or containerised environments, using technologies such as Kubernetes, AWS or equivalent platforms.
• Strong German communication skills at C1 level or Native.
• An ownership mindset, sound technical judgement and the ability to turn loosely defined requirements into working software.

Additional skills that would be advantageous:

• Strong SQL skills, including database modelling, API design and query optimisation.
• Experience deploying AI or machine learning models into production and evaluating their practical limitations.
• A product-oriented approach to engineering, with the ability to work directly with users and understand their needs.
• Familiarity with event-driven systems, message brokers such as Kafka or RabbitMQ, or similar distributed architectures.
• Experience using AI-assisted coding tools such as Claude Code, Codex or comparable solutions.

You do not need to have worked with every technology listed. Strong backend engineering fundamentals and hands-on experience building agentic AI solutions are the priorities.

Ready to shape how intelligent agents transform complex workflows into scalable software? If you enjoy hands-on engineering, technical ownership and building AI systems that solve real operational challenges, we’d love to hear from you.