Full-Time

AI Engineer

Deeplight

Deeplight

11-50 employees

Coherent, compact lasers for space LiDAR

No salary listed

Abu Dhabi - United Arab Emirates

Hybrid

Hybrid work arrangement in Abu Dhabi.

Bachelor's, Master's

Category
AI & Machine Learning (1)
Required Skills
LLM
Power BI
Kubernetes
MLOps
Microsoft Azure
Python
NoSQL
Data Science
SQL
Machine Learning
Postgres
MLflow
Docker
RAG
Tableau
Observability
MongoDB
REST APIs
LangChain
DevOps

Get referred to Deeplight

See people who can refer or advise you

Requirements
  • A bachelor's degree in a quantitative field such as Computer Science, AI, Software Engineering, or Mathematics is required.
  • Five or more years of experience in AI/ML engineering or large-scale digital product engineering is required, ideally within a major digital transformation or digital-only bank project.
  • Proven experience deploying large-scale AI/ML solutions within a regulated environment is required.
  • Hands-on experience with MLOps tools including MLflow and CI/CD, and data visualization platforms including Tableau and Power BI, is required.
  • Expert-level Python programming for machine learning, artificial intelligence engineering, and Generative AI applications is required.
  • Deep knowledge of transformer models, multimodal architectures, and agentic AI systems, including multi-agent frameworks and human-in-the-loop systems, is required.
  • Expertise in vector databases such as Pinecone, Weaviate, and Milvus; knowledge graphs; and SQL and NoSQL systems such as PostgreSQL and MongoDB is required.
  • Proficiency in Azure AI Foundry or AWS SageMaker, including serverless deployments with Lambda or Azure Functions and containerization with Docker and Kubernetes, is required.
  • Deep understanding of large language model evaluation metrics and observability frameworks for monitoring agentic workflows is required.
Responsibilities
  • Build and scale next-generation conversational AI platforms for voice and chat using large-language-model-based agents and multi-agent orchestration frameworks.
  • Develop automation agents for critical banking functions including know-your-customer, compliance, risk monitoring, and customer service.
  • Implement retrieval-augmented generation pipelines using vector databases and knowledge graphs to power enterprise search.
  • Design and refine prompts to optimize large language model performance and apply parameter-efficient fine-tuning techniques.
  • Operationalize embedding models and custom machine learning solutions by building APIs, software development kits, and continuous integration and continuous delivery pipelines with automated retraining and drift management.
  • Establish engineering standards for model fairness, security, explainability, and disaster recovery to ensure operational resilience.
  • Identify opportunities to use data to solve complex business problems across functional teams and translate analytics into business value.
  • Present technical decisions and proposed solutions to high-level stakeholders and articulate their business rationale.
Desired Qualifications
  • A master's degree in AI or machine learning is preferred.
  • Experience in banking or financial services is preferred.
  • Ability to quickly build and test AI prototypes while maintaining enterprise-grade security and compliance.
  • Active engagement in the AI community and experience evaluating emerging tools such as LangChain, CrewAI, and Bedrock.
  • Strong communication skills for bridging technical requirements and business objectives with non-technical stakeholders.
  • Understanding of emerging trends such as reinforcement learning from human feedback and multimodal AI.

DeepLight develops highly coherent, compact lasers tailored for space research and laser technology applications. Its lasers span the visible and near-infrared spectrum and feature exceptional frequency agility to support advanced applications, including long-range coherent LiDAR. The products rely on a combination of compact light sources and integrated photonics, including MEMS-based ultrafast acousto-optic modulation, enabling fast, precise control of laser light. Unlike many competitors, DeepLight emphasizes tight coherence, high linearity, and integration with MEMS for rapid modulation, supported by collaborations with major space agencies like the European Space Agency and revenue streams from technology sales and tech-transfer funding. The company's goal is to commercialize its laser technology for space research, communication protocols, and fibre sensing, accelerating the deployment of next-generation laser sources in science and industry.

Company Size

11-50

Company Stage

Debt Financing

Total Funding

$90K

Headquarters

Saint-Sulpice, Switzerland

Founded

2021

Get referred to Deeplight

See people who can refer or advise you

Simplify Jobs

Simplify's Take

What believers are saying

  • In 2026, Deeplight secured Horizon Europe EIC grants for FORTE, AGILIGHT, and CombTools.
  • April 2026 partner validation for DLT-FT-1550 supports credibility in fiber sensing and QKD.
  • EPFL and FIT backing, plus 2025-2026 product updates, reduce technical and fundraising risk.

What critics are saying

  • Deeplight remains 11-50 employees in 2026, with fragile execution capacity.
  • The company still needs certifications and first-customer contracts after FIT’s 2023 seed loan.
  • If partner validation stalls in 2026, Deeplight becomes a grants-dependent lab, not a business.

What makes Deeplight unique

  • Deeplight’s April 2026 DLT-FT-1550 hits below 5 Hz intrinsic linewidth.
  • Its PIC-based lasers combine 2 GHz tuning with MHz-bandwidth actuation, unusual for startup hardware.
  • DeepLight targets QKD, fiber sensing, LiDAR, and space telecom from Swiss-German labs.

Help us improve and share your feedback! Did you find this helpful?

Benefits

Health Insurance

Professional Development Budget

Performance Bonus

Growth & Insights and Company News

Headcount

6 month growth

-5%

1 year growth

-5%

2 year growth

0%
StartupTicker
May 7th, 2024
Innosuisse Funded 45 Start-Up Innovation Projects

Since the beginning of 2023 Innosuisse has been able to directly support innovation projects from start-ups before they go to market. Up to now 45 start-ups received funding – check the full list.With the Start-up innovation project, Innosuisse directly funds high potential science-based projects initiated by Swiss start-ups, giving them a financial boost to propel their market entry. The innovation projects must be based on application-oriented research and prepare the start-ups for entering the market for the first time. In the case of these projects, only the start-up is supported. A research partner is not necessary. Furthermore, eligible companies must be headquartered in Switzerland, were established not more than 5 years ago and have less than 50 full-time equivalent employees.Since the start in 2023, 45 companies received direct funding for their science-based projects

StartupTicker
Nov 21st, 2023
Deeplight and ProSeed get FIT Funding boost

Deeplight: affordable high-end lasers for mass marketsIn collaboration with EPFL's Laboratory of Photonic Integrated Circuits and Quantum Measurements, Deeplight is at the forefront of the development of innovative lasers based on hybrid photonic integrated circuit technology. These lasers are essential for applications such as satellite-to-satellite telecommunication, LiDAR remote sensing in the automotive sector, and greenhouse gas detection.
 
These lasers feature the highest spectral purity on the market enabling optical detection applications with much greater sensitivity than today.
Deeplight has already signed licensing and collaboration agreements with EPFL, as well as development contracts with leading industrial companies.
 
The CHF 100,000 Tech Seed loan from FIT will enable Deeplight to strengthen the robustness of its lasers and achieve key milestones, including obtaining the relevant laser certifications and concluding contracts with its first customers.
 
ProSeed gives a second life to brewing by-productsProSeed's mission is to transform the Swiss food industry by upcycling brewing by-products into valuable raw materials. Using an innovative production unit, ProSeed offers breweries a hassle-free solution for transforming their by-products into food-grade raw materials, generating additional revenue while significantly reducing waste.
 
ProSeed's business model involves setting up processing units in breweries to convert brewers' residual grain into barley flakes, which it then sells to its network of ingredient manufacturers. The three-step process has been patented and the start-up has already signed letters of intent with breweries and food manufacturers. By teaming up with a major bakery, ProSeed sold over 8,500 baguettes made from barley flake-based protein flour in three months.
 
With the CHF 100,000 FoodTech loan from FIT, ProSeed will optimise its technology and increase its production capacity, with the aim of generating revenue by 2025.(Press release)