Agentic Engineering Expert (m/w/d), Krakow at TOWA - the digital growth company

Krakow, Lesser Poland Voivodeship, Poland -

Full Time

Start Date

Immediate

Expiry Date

21 Sep, 26

Salary

230000.0

Posted On

23 Jun, 26

Experience

5 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Claude Code, Agentic Workflows, MCP Servers, TypeScript, Python, CI/CD, Context Engineering, LLM Evaluation, Software Engineering, Prompt Engineering, Multi-agent Systems, Spec-driven Development

Industry

Business Consulting and Services

Description

What we stand for: “We drive digital transformation end-to-end – from strategy to execution. Creating measurable progress and real impact for our clients.” Standing still? Not us. We believe: Everything is BETA – and that’s exactly where the opportunity for progress lies. While others are still debating AI strategy, we’re rebuilding our delivery to be AI-native – with specs, verification gates, and agents across the entire development process. And now it’s your turn: We’re looking for the person who will accelerate this transformation across our engineering teams. Not as someone who just delivers concepts, but hands-on – as someone who lives it and pulls others along. Your Mission: Your mission: You make our development teams faster and better by building agentic workflows, setting standards, and multiplying knowledge. You’re more than a technical expert here – you’re a change agent: a frontrunner and innovator who leads the way in projects, lives the new way of working, and brings the teams along. One clarification up front: this is not an AI/ML role. You don’t build models – you build software faster by orchestrating AI agents and designing the environments in which they work reliably. Claude expertise, workflows & agent teams You’re our Claude expert – the deepest knowledge of the Anthropic ecosystem in the house, from Claude Code (CLAUDE.md, subagents, hooks, skills, plan mode, headless/SDK in CI/CD) to the Agent SDK You build agentic workflows and agent teams: multi-agent setups that research, develop, and review in parallel – and work reliably in day-to-day project life You build and operate MCP servers that make our tools and data sources usable for agents You establish spec-driven development: standardised inputs, clear verification gates, auditable agent output Quality & verification You develop eval harnesses and testing procedures that make AI-generated code measurable and trustworthy You treat context engineering as a discipline: the smallest set of high-signal tokens, not the longest prompt You review AI output critically – “trust then verify” is your default stance, not a slogan Change & enablement You work as a frontrunner directly within the project teams: you show on a real client project what’s possible with agents – and then enable the teams to do it themselves You drive change as a change agent: you inspire rather than preach, bring sceptics along, and clear away obstacles in both processes and mindsets You build reusable assets: plugins, skill libraries, prompt patterns, runbooks You watch the ecosystem (Anthropic, the MCP standard, tooling) as an innovator, try new things early, and translate them into usable practice for TOWA You can score points with the following: Agentic engineering experience You’ve built productively with LLMs – not just watched tutorials. The litmus test: you can describe an evaluation that you designed yourself Deep hands-on experience with Claude Code or comparable agentic coding tools: you know when agents shine, when they fail, and how to tell the difference Experience building MCP servers, custom tools, or agent integrations Software engineering foundation A solid engineering background (e.g., TypeScript, Python) – you’re a developer who works with agents, not a prompt author without a codebase An understanding of CI/CD, testing, and delivery processes that agents need to fit into The LLM fundamentals are second nature: tokens, context windows, model tiers, cost/latency trade-offs Mindset & communication Learning velocity: the tools change every month – you grow at least as fast Ownership: you stand by every commit with your name on it, even when an agent wrote it Change-agent qualities: you meet people where they are, lead by example, and persevere when change meets resistance You can explain, enable, and convince – within the team and with clients, in German or English Nice to have Experience in an agency or consulting environment with parallel projects Hands-on experience with eval frameworks, LLM-as-a-judge setups, or agent observability Contributions to the community: open source, talks, blog posts What we need from you: Mission started?! Then here’s what we need from you: Your CV A concrete example of how you’ve worked with agents (setup, workflow, or asset – format doesn’t matter) What compensation do you have in mind for your performance Your notice period or possible start date Why us? We believe: those who give a lot should also get a lot back. Pure team spirit: collegial teams with real cohesion, a willingness to help, and a healthy dose of humour Transparent structures: a genuine open-door mentality – communication as equals is a given Open vacation policy: on top of your statutory holiday entitlement, our “Above & Beyond” scheme lets you treat yourself to additional days off Hybrid working: home office and flexibility around our locations – being close to the team and clients matters to us Flexible working hours: a wide flexitime framework with reliability for client and team meetings Growth guaranteed: an individual development budget for workshops, certifications, or coaching Career paths: clear level structures for your development – an expert track or a leadership journey Mobility & events: a subsidy for public transport or the Deutschlandticket, regular team events, trade fairs & conventions Loft offices: centrally located, modern equipped – with space for exchange, ideas, and of course snacks & drinks Rewards: a bonus system for your personal and entrepreneurial successes This position sits at a senior level within TOWA's Solution Cluster at the Krakow location. For this role, we offer two contract models: B2B contract: PLN 95–155 per hour (net) Employment contract (umowa o pracę): PLN 140,000–230,000 gross per year The stated range covers all salary components and is to be understood as a target salary (e.g., including variable components or further benefits, where applicable). Should we work together, we'll put together an individual offer for you – your grading within the range is based on your skills, your area of responsibility, and your experience, never on your previous salary. In line with the EU Pay Transparency Directive (2023/970) and the country-specific equal treatment laws, we publish the salary range for every role and do not ask candidates about their current or previous salary. Our criteria for setting pay – cluster, level, skills, area of responsibility, and experience – apply uniformly across the entire company. Throughout the whole process, you can request further information about the salary range for this role or about our compensation policy in writing at any time. Equal opportunity TOWA hires based on skills and impact. We welcome applications from all qualified people – regardless of gender, age, background, origin, religion, sexual orientation, or disability. The job title applies to all genders. Vorgesetzter Your lead is Matthias Frank - click here to get to know him. About us As an internationally award-winning digital agency based in Bregenz, Vienna, Hamburg, Munich, Dortmund and soon Krakow, we support medium-sized companies in the conception & realisation of their digitalisation strategy. We work on sophisticated user interfaces, web applications, high-traffic portals, digital design projects, online marketing campaigns as well as data visualisation and analysis. Leading technology companies such as Spryker, Storyblok, Pimcore, Salesforce and HubSpot are our partners here.

Responsibilities

Build and orchestrate agentic workflows and multi-agent teams to accelerate software development processes. Act as a change agent to implement AI-native delivery standards and enable engineering teams through reusable assets and mentorship.