ML Engineer

About Alaan

Alaan is the Middle East’s first AI-powered spend management platform, built to help businesses save time and money.

Our all-in-one solution combines smart corporate cards, real-time expense tracking, AI-powered automation, seamless accounting integrations, and deep financial insights- designed to simplify finance operations and maximize control over company spend.

Founded in 2022, Alaan is already the trusted partner of over 3000 leading businesses across the UAE and KSA, including G42, Careem, McDonald’s, Tabby, Al Barari, Rove Hotels, Rivoli, and CarSwitch. Together, our customers have saved over AED 100 million with Alaan.

In just Four years, Alaan has been named to the Forbes Fintech 50 list, UAE Future 100 list, selected as an Endeavor company, and ranked the No. 1 spend management platform in EMEA by G2.

Alaan is also backed by Y Combinator and top global investors, including Peak XV (formerly Sequoia India), Y Combinator, 468 Capital, and 885 Capital, among others, and built by a world-class team with experience from McKinsey, BCG, Goldman Sachs, Careem, Rippling, and other high-growth companies. This creates an environment where employees learn from top-tier talent while building at startup speed.

We’re not just building software. We’re reimagining how finance works for modern businesses across the region.

*Alaan is a fintech company, not a bank. Business bank accounts are issued by Ruya.

About the Role

We are looking for a Machine Learning Engineer to build the intelligence behind Alaan's expense automation: systems that understand financial documents, transactions and language, act on them with minimal human input and get better with every interaction. You will work on open-ended, high-ambiguity problems at the intersection of deep learning, large language models and real-world financial data, owning them from definition to production and measurable impact.

What you'll do

• Translate ambiguous business goals into well-posed machine learning problems with clear success metrics, and shape the technical direction of how we solve them.

• Design and train deep learning and multimodal models that understand transactions, documents and free text and produce structured, reliable outputs.

• Build self-improving systems that learn continuously from user feedback, explicit and implicit, using RLHF, preference optimization, reward modeling and contrastive learning.

• Fine-tune and adapt large language and embedding models for domain-specific understanding, balancing quality, latency and cost.

• Architect ML systems end to end: data and feedback pipelines, training and evaluation infrastructure, real-time serving, monitoring and safe rollout.

• Set up rigorous evaluation through offline benchmarks, online experiments and guardrails that connect model quality to customer and business outcomes.

• Apply anomaly detection and risk modelling to surface fraud, policy breaches and data-quality issues.

• Track the state of the art, assess new methods critically, and bring the ones that work into production, in close partnership with product and engineering

What we are looking for

• Bachelor's degree in Computer Science, Statistics, Mathematics or a related field, or equivalent practical experience.

• 2 to 4 years building and deploying machine learning systems in production with demonstrated product impact, ideally in fintech or payments.

• Deep understanding of modern deep learning, especially transformer architectures, with experience designing, training and debugging models beyond off-the-shelf use.

• Hands-on experience fine-tuning large language and embedding models (supervised fine-tuning, parameter-efficient methods such as LoRA, contrastive learning).

• Strong fundamentals in machine learning, statistics and optimisation, including classical methods, and the judgement to choose the simplest approach that works.

• Experience designing production ML systems; proficiency in Python, PyTorch and SQL, and cloud platforms(AWS preferred).

• Ability to break down ambiguous problems, form hypotheses, run disciplined experiments and explain trade-offs to technical and non-technical audiences.

Nice to have

• Master's or PhD in machine learning or a related field, or publications at venues such as NeurIPS, ICML, ACL or KDD.

• Production experience with RLHF, DPO, reward modelling or other learning-from-feedback methods.

• Experience with document understanding and vision-language models.

• Experience building conversational or agentic LLM systems, including retrieval and tool use.

• Background in fraud detection, risk modelling or transaction monitoring.

• Experience with continual, online or active learning, or cold-start problems at scale.

What's in it for you

• Contribute to building the Middle East’s most beloved fintech brand from the ground up

• Benefit from a role with significant ownership and accountability

• Thrive in a flexible hybrid culture with ample work-life balance

• Participate in exciting offsite events

• Competitive salary and equity

• Enjoy additional perks like travel allowances, gym memberships, and more