AI Engineer - M&A Services

Duely BV; Oost-Vlaanderen; Remote; EUR 70,000 - 100,000; Full time

Benefits offered by this job

Equity

Free lunch

On-site Ghent office

Challenging projects

Job summary

Duely BV in Ghent is seeking a hands-on AI systems engineer to build and run the AI workloads behind live M&A deliverables.

You will work with legal experts, write production workflows, measure quality and cost, and ship improvements for future deals. This on-site role offers direct impact and fast feedback.

Qualifications

• Strong Python programming and production code experience.
• Experience with LLM-powered systems and retrieval concepts.
• Ability to explain technical trade-offs to non-engineers.

Responsibilities

• Ship to production: build and deploy LLM-powered extraction and document analysis workflows.
• Measure quality, turnaround time, and cost of outputs.
• Close the loop by finding root causes and shipping fixes with tests.
• Make it scale: process thousands of documents per deal.
• Work closely with lawyers to translate requirements into software.

Skills

Python

LLM systems

Retrieval

Agents

Eval pipelines

Harness engineering

Job description

Build and use the AI systems behind live M&A services, side by side with the legal experts responsible for the work.

Clients instruct us as they would instruct a law firm and receive the finished work: data room preparation, gap analyses and full legal due diligence reports. Our AI engine does the heavy lifting, our legal experts direct and sign off on every deliverable, and we carry full responsibility for the output. We price per deliverable, not per billable hour.

Founded in Ghent by Michiel Van Lerbeirghe (former lawyer and GC) and Michiel De Koninck (AI engineer and legal-tech lead), we have raised €1.1 million in venture capital, and our technology is integrated directly into Sterling's global VDR platform. We are still small enough to fit in one room, so you will directly shape how the company is built.

Our values

• Excellence in execution. An eye for detail in deliverables, both external and internal. Crossing your t's and dotting your i's.
• Everybody sales. Everyone has the growth of the business in mind.
• Take ownership. Our culture allows everyone to speak their mind: do so, and take ownership where you know you can make a difference.
• Be brutally honest: no ego. To grow, as a team and individually, we need to give each other constant and useful feedback, respectfully.

About the role

You build and run the AI systems behind each deliverable. On live deals, delivery comes first: you write and adapt production workflows, debug failures, and work alongside the legal expert until the output meets our standard, on time.

Each deal feeds directly back into the codebase. You measure quality, turnaround time and cost, trace extra review and delays back to their cause, and ship reusable improvements for every deal that follows.

This is hands-on LLM application engineering, not model training. Owning that delivery + improvement loop is your job and it's crucial to the success of Duely.

Responsibilities

• Ship to production. Build and deploy LLM-powered extraction, document analysis and agentic workflows used on real deals.
• Measure everything. You can't improve what you can't measure. Build the evaluation pipelines that tell us whether output is good enough to put in front of a client.
• Close the loop. Find the root cause of weak output, ship the fix and add the eval that catches it next time.
• Make it scale. Process thousands of documents per deal, reliably and without the cost running away.
• Sit with the lawyers. Work closely with the legal experts who use your work. Learn how they work and think, and turn the repeatable parts into software.

Requirements

• You ship production code. You write strong Python, have built LLM-powered systems and understand retrieval, agents, evals and harness engineering. Serious side projects count.
• You have good engineering judgement. You know when to move fast, when to slow down and what needs to be reliable.
• You can talk to both sides. Explain a technical trade-off to a lawyer and a legal nuance to an engineer.
• You enjoy messy, unfamiliar problems. You like learning directly from domain experts.
• You want to be in the room. This role is on‑site in Ghent.
• Bonus: Experience in legal tech, M&A or another high‑stakes, document‑heavy field, or with observability, data privacy, security or cloud infrastructure.

Why Duely

• Meaningful equity, early. You are joining early enough that your slice matters more than rounding to zero.
• The shortest feedback loop you will find. Real deals arrive every week, and the legal expert reviewing your work sits next to you. You know within days whether it held up.
• A profession being rebuilt. M&A legal work is being reinvented right now. Help decide what it looks like in an AI‑native world.
• Free lunch. Oh, you thought there was no such thing? There is at Duely. All meals at our office are paid for because we love to fuel your engine. MVL's tuna salad is quite legendary, Simon's yoghurt is quite dairy.

The team posing in their plant-filled (bit problematic) kitchen. LTR: Lucas, Simon, MVL, Arne, MDK