Skip to Content

Optimization Engineer

--Josep Fanals Batllori--

About eRoots


Constituted in 2022, eRoots is an innovative software startup based in Barcelona dedicated to shaping power systems analysis through cutting-edge technology. We focus on developing advanced solutions that enhance renewables integration, grid stability, optimal operation and optimal planning using state-of-the-art algorithms and machine learning techniques. Our team of 15 engineers is composed of passionate experts aiming to make a significant impact on modern power systems. 

Job description


Power grids are large infrastructures that require careful planning to ensure that electricity is delivered to all consumers. This planning stage is complex and involves many variables, such as the location of power plants, the capacity of transmission lines, among others. Bear in mind the decisions made here have both a technical and economical component that companies want to optimize for. Traditionally, this planning has been done by so-called human experts who use their knowledge and experience to supposedly make the best decisions. However, this process is time-consuming and error-prone, and it is becoming increasingly difficult considering the numerous assets to take into account.

The rapid advances in machine-learning techniques, coupled with the availability of large amounts of data and computing power, have opened up new possibilities for planning power grids. In particular, reinforcement learning (RL) has shown great promise in solving complex optimization problems. This is where you come in. We are looking for someone who can help us develop and implement RL to optimize the planning of power grids. With a single click of a button, customers should get the best plan for their power grid, taking into account all the variables and constraints. 

Responsibilities


  • Develop and implement reinforcement learning models to optimize transmission system planning.
  • Experiment with procedural generation techniques and define policies to orient the RL agent.
  • Collaborate with cross-functional teams to integrate these models into our software solutions (see GridCal).
  • Conduct rigorous unit and integration tests to validate the algorithms, ensuring code robustness.
  • Stay abreast of the latest advancements in AI and machine learning, applying them to enhance our products.
  • Document and present model development processes, outcomes, and insights to both technical and non-technical stakeholders.

It is worth reminding you that a startup is a fast-paced environment. This most likely means your work can significantly pivot and be reoriented if need arises. 

Requirements


We value competency and loyalty above everything else. The points below are mostly nice-to-have rather than rigid musts you have to meet:

  • Experience in optimization algorithms, ideally within the power systems sector.
  • Basic ideas on the elements that constitute an electrical grid and how it is operated.
  • Knowledge on steady-state calculations for power systems. Mainly power flows and short-circuits.
  • Solid programming skills in Python and familiarity with machine learning frameworks.
  • Strong analytical and problem-solving abilities.
  • Ability to work collaboratively in a fast-paced, agile environment.
  • Decent communication and presentation skills.
  • A proper level of English is mandatory. Catalan and Spanish are valued but not required.

What we offer


  • A competitive salary and comprehensive benefits package.
  • Flexible working hours and hybrid work.
  • Opportunities for professional development and career growth.
  • A collaborative, inclusive, and innovative work environment.
  • The chance to be part of an Barcelona-based software company with a cohesive team.

Recruitment process


  1. An initial 15-minutes conversation to quickly get a glimpse of your interests and understand if you could be a good fit to the company.
  2. A deep technical interview with our CEO. Expect a dialogue in which we go over power systems modelling, algorithms, numerical methods, and coding.
  3. An optional final interview with some of our co-founders and leading team members.