This page was automatically translated and may contain errors. View in English.
a

AI Systems Engineer for Time Series Forecasting, Optimization, and LLM-based Coordination (m/f/d)

adjusted flow

Karlsruhe, Baden-Württemberg, Germany · മുഴുവൻ സമയവും

അപേക്ഷിക്കുന്ന ആദ്യയാളാകൂ

അനുഭവം
2+ yrs
ശമ്പളം
ഓപ്പണിംഗുകൾ
1
പോസ്റ്റ് ചെയ്തു
2 മണിക്കൂർ മുമ്പ്
Work mode
ഓഫീസിൽ
Eligibility
Applicants who are motivated to help build an early-stage product, take ownership, and contribute ideas are encouraged to apply. The company especially values people with hands-on, execution-oriented experience. A résumé is welcome, and a short note about your motivation and relevant projects is pr…
Resume
Required to apply

Where you'll work

ജോലി വിവരണം

About the Company and Mission

adjusted flow is an early-stage startup based in Karlsruhe that is building a software platform for energy consultants and energy-intensive manufacturers. The platform is designed to help customers make better use of flexibility on-site, balancing generation and consumption in a way that supports the energy transition where it matters most: industrial operations.

The company is still in the early phases of product development and is laying the foundation for a scalable solution. It is looking for an AI and Forecasting Developer (m/f/d) to help build and evolve the platform.

Role Overview

You will help shape the AI and decision-making core of EOIS for industrial energy optimization. This work brings together probabilistic forecasting, uncertainty handling, optimization coupling, and carefully bounded supervisory intelligence for real-world industrial systems.

What You Will Work On

  • Build and refine forecasting models for industrial electricity demand, PV generation, and other energy-related signals.
  • Expand deterministic forecast services into probabilistic outputs such as quantiles, confidence scores, and well-calibrated uncertainty intervals.
  • Measure forecast performance on industrial data that is noisy, incomplete, and non-stationary.
  • Define how forecasts and uncertainty estimates feed downstream optimization, especially for flexible assets such as battery storage.
  • Help develop the supervisory intelligence layer that reads system state, forecast quality, and service health.
  • Create and improve controlled decision rules, including re-forecasting, switching operating modes, suppressing low-confidence outputs, and escalating to a human when needed.
  • Build structured AI workflows and integrate them into production Python services, gRPC interfaces, and the live EOIS platform.
  • Support validation, benchmarking, and technical interpretation of results in actual customer environments.

What the Company Is Looking For

The team is not seeking a perfect checklist. Instead, it values people who think independently, ask questions, and are motivated to build something meaningful.

  • You take initiative instead of waiting for detailed instructions.
  • You notice when something can be improved and bring your own ideas.
  • You are comfortable challenging assumptions when you believe there is a better solution.
  • You are willing to take responsibility even when the setup is still evolving.

From a technical perspective, experience in several of the following areas is expected:

  • Python development
  • Forecasting for industrial or energy-related time series
  • Probabilistic forecasting techniques such as conformal prediction, quantile regression, ensemble methods, or probabilistic boosting
  • Calibration and uncertainty assessment
  • Understanding of LP/MILP or similar optimization methods
  • Practical familiarity with Pyomo, PuLP, or comparable optimization tools
  • Designing model outputs for downstream decision systems
  • LLM application development with structured outputs
  • Defining bounded decision and action spaces
  • Knowledge of industrial energy systems, especially PV, BESS, and load, with hydro experience being a plus

Helpful additional experience includes:

  • pvlib or comparable PV modelling tools
  • Open-source LLM serving stacks such as vLLM, Ollama, or llama.cpp server
  • Replay testing or simulation frameworks
  • Explainability, auditability, or human-in-the-loop AI systems

Experience Expectations

  • Ideally, you have at least 2 years of overall work experience.
  • At least 1 year should come from a small company, startup, or industrial research lab.
  • Practical experience in execution-focused, non-state-funded environments is preferred.
  • Backgrounds limited to universities, publicly funded academic research, or very large highly structured corporations are generally not sufficient on their own for this role.

Benefits and Working Environment

The company does not offer a long list of standard corporate perks, but it does offer the chance to make a real impact. You can expect coffee, snacks, a strong team, and openness to new ideas.

  • High responsibility from day one
  • A steep learning curve and exposure to unfamiliar challenges
  • Direct influence on the product and important decisions
  • The chance to work on a meaningful challenge in the industrial energy transition
  • Close collaboration and fast decision-making
  • Potential to grow into a key long-term role in the company

Application Notes

The company is still building its foundation, so things may not always be perfectly structured. That is also what makes the opportunity impactful for someone who wants to help build, take ownership, and improve the way things work.

A résumé is welcome, but the team would especially appreciate a short note explaining your motivation for the role and the projects in which you applied the technical skills relevant to this position.

Start Date

The preferred start date is 1 June.

മറുപടി വേണമെങ്കിൽ അത് വിടുക — ഞങ്ങൾ അത് മറ്റൊന്നിനും ഉപയോഗിക്കില്ല.

ബ്രൗസ് ചെയ്യാൻ ക്ലിക്ക് ചെയ്യുക, വലിച്ചിടുക, അല്ലെങ്കിൽ പേസ്റ്റ് ഒരു സ്ക്രീൻഷോട്ട്

PNG, JPG, GIF, MP4, WebM, MOV · പരമാവധി 20MB ഓരോന്നും · 5 ഫയലുകൾ വരെ