Forward Deployed Engineer

Type of contract: full-time or trade license

Place of work: Kosice or remote

Start date: let us know what time suits you best

Compensation: from €25/hour up to €46/hour

About the role

At Sudolabs, we embed with one client at a time and put AI to work on their hardest, most valuable problems: automating the manual, expert-driven work that runs an enterprise, modernising the legacy systems it depends on, helping their engineers adopt AI in how they build, and standing up entirely new AI products from a blank page. We go deeper into the problem than the client has, then build the solution that actually works, in production, with the best tools out there.

A Forward Deployed Engineer (FDE) sits at the sharp end of that model. You're the engineer who lands inside a client's world of legacy systems, spreadsheets, month-end chaos, and compliance constraints, and turns it into shipped, production software. You lead discovery, architect the solution, write the code, put it in front of real users, and iterate until it's genuinely useful. You're equal parts builder, problem-framer, and trusted technical partner.

This is the role a16z called "the hottest job in tech", and it's the role behind our highest-impact engagements: automating a finance team's month-end close, building an AI "conversion factory" for a legacy data migration, standing up an AI coaching platform from a blank page. If you want to see your work land in production and change how a business runs in weeks, not quarters, this is for you.

We're a services company. Your "platform" is the client's real environment, and the reusable harnesses, patterns, and playbooks you leave behind so the next engagement starts faster.

The kind of work you'll do

Our engagements recur in a handful of shapes, and you'll move between them:

• AI automation in enterprises: replacing manual, expert-driven work (month-end close, compliance checks, large-scale data migration and reconciliation) with agents and validation harnesses that give full coverage, prove their output against ground truth, and leave humans on the edge cases.
• Legacy system rewrite & modernisation: taking an aging, business-critical, often undocumented system and rebuilding it on a modern stack. You'll use AI to accelerate reverse-engineering, code translation, and test generation, and reconcile behavior against the old system so the cutover is safe rather than a leap of faith.
• AI-enhanced SDLC: embedding inside a client's own engineering team to help them adopt AI: standing up tooling, agent harnesses, evals, and workflows, and building those harnesses around their product with them, so the team ships faster and is more capable after we leave, not dependent on us.
• AI products from a blank page: standing up a new AI-powered product (agents, voice, multi-signal scoring) from discovery to production.

What you will do

• Embed with clients from kickoff through go-live and beyond: you're the technical face of the engagement, not a spec-taker.
• Lead discovery on messy, real-world problems: map manual workflows, reverse-engineer undocumented legacy systems, and surface the assumptions and questions that de-risk the build.
• Improve the process, don't just automate the broken one: figure out what should happen before you make a machine do what currently happens.
• Architect and build custom AI solutions using best-in-class tools: LLM agents, data pipelines, validation/reconciliation harnesses, exception-handling workflows, and human-in-the-loop review UIs.
• Ship fast and iterate on real data: our engagements run in tight prototype sprints, so you'll get real solutions in front of real users quickly.
• Design for trust in high-stakes, often regulated domains: full coverage, humans on exceptions, auditable outputs, and evals that prove the thing works.
• Build relationships with client stakeholders: from controllers and ops leads to CTOs and execs, translating fluidly between business value and technical reality, and inspiring confidence in what you're building.
• Capture reusable patterns, harnesses, and playbooks so knowledge compounds and every next engagement is faster. This is how a services company builds leverage.

What we're looking for

• Strong software engineering fundamentals: you can design and ship production systems, not just prototypes, and you're comfortable across the full stack (TypeScript / Node / React and Python).
• Hands-on with modern AI tooling: LLMs from concept to production, agent frameworks, prompt/context engineering, RAG strategies, and evaluation frameworks to assess whether AI output is actually good enough.
• Systems thinking and engineering judgment: you can look at a tangled process and see where the leverage is, and you know when not to build.
• Client-facing skills: you can lead a working session, manage stakeholders, explain a complex trade-off to a non-technical exec, and build trust through usefulness.
• High agency and a bias to ship: you're comfortable with ambiguity and changing requirements, and you own outcomes end to end (Palantir describes the mindset as "a startup CTO", and that's the energy).
• Commercial awareness: you understand you're delivering business value on a client's budget, and you keep scope honest.
• Fluent English and comfort working directly with clients across time zones (we work mostly with US companies).

Nice to have

• Prior experience in a forward-deployed, solutions, or applied-AI engineering role with real customer exposure.
• Domain depth in one of our recurring problem spaces: finance/accounting operations, insurance/insurtech, data migration and reconciliation, legacy modernisation, or developer tooling / AI-assisted SDLC.
• Data engineering skills: building reliable pipelines and wrangling messy, multi-format, large-scale data (SQL, ETL, warehouses).
• Data science / ML background: analytics, modelling, and evaluation beyond off-the-shelf LLMs.
• DevOps/infra comfort (AWS, serverless, queues) and enough frontend chops to build a clean human-in-the-loop review interface.
• Experience hardening production systems on top of alpha/unstable third-party AI services.
• A track record of turning one-off client work into reusable assets.

How we work

• Remote-first - most of our engagements are delivered remotely, and there's no mandatory desk time, so work fits seamlessly into your life. We have a dog-friendly HQ (and a Sudo apartment) in Košice if you want it, but you don't need to be there.
• Occasional travel - some engagements benefit from being on-site for kickoff or discovery (typically the first week or two). It's the exception, not the rule, and it's planned with you. This is not a "live on a plane" job.
• Small pods - you'll work in a compact, cross-functional team (typically ~3-5: delivery/product lead, AI/data engineers, designer) focused on a single client, using rapid prototyping and fast delivery as the default.

What's in it for you

• A front-row seat to applied AI at the frontier: real production systems, real business impact, and the variety of a new hard problem every few months.
• Permanent contract or trade license — choose what fits you
• Work from our office in Kosice or from wherever you like
• 20 paid days/year in your work time to learn new things
• 31 paid days/year time off
• Flexible working hours
• Internal tech talks on a regular basis
• Hardware package — MacBook or anything you need
• Team buildings and company retreats
• Sudo apartment in Kosice for remote colleagues
• Brand new pet-friendly office in the city center with parking space