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

Senior Data Scientist – Python, Machine Learning & Signal Processing

DATATRONiQ

Berlin, Germany পূর্ণকালীন

প্রথম আবেদনকারী হোন।

অভিজ্ঞতা
৩+ বছর
বেতন
শূন্যপদ
1
পোস্ট করা হয়েছে
১ ঘন্টা আগে
কাজের ধরণ
অফিসে
শিক্ষা
Bachelor's degree or higher in Data Science, Computer Science, Mathematics, Physics or similar
জীবনবৃত্তান্ত
আবেদন করা আবশ্যক

যেখানে আপনি কাজ করবেন

কাজের বিবরণ

Company Overview

DATATRONiQ is a German deep-tech startup specializing in Industrial IoT and Edge AI solutions. As a Senior Data Scientist, you will develop and train machine learning models based on real-world machine and sensor data, serving clients ranging from mid-sized manufacturers to large DAX-listed corporations worldwide. Your work will directly impact operational decisions, such as preventing unplanned downtime and improving production quality in real time.

Role Overview

You will oversee the full data science lifecycle, including data exploration, feature engineering, model training, deployment, and validation, tailored to client environments—whether on edge gateways, on-premises servers, or cloud platforms. Time series data comes from industrial controls using OPC-UA and MQTT protocols, requiring robust signal processing on noisy sensor signals rather than just table-based transformations. Your models will be quantized, exported as ONNX, and deployed where production needs dictate, demanding careful consideration of model architecture, latency, and memory usage.

Technologies involved include Python, PyTorch or scikit-learn, ONNX for edge deployment, and common MLOps tools supporting versioning and reproducibility. Team collaboration involves regular code reviews, pair programming, and weekly sessions showcasing new tools and discoveries. Close partnership with data engineers and backend developers ensures reliable model integration into production pipelines, minimizing notebook-only prototypes.

Key Responsibilities

  • Train machine learning models focused on anomaly detection, predictive maintenance, and quality monitoring validated against actual production and industrial control time series data.
  • Perform feature engineering on noisy machine and sensor signals, applying signal processing and filtering techniques to OPC-UA, MQTT, and MES-exported data.
  • Deploy models in customer environments—edge gateways, on-prem servers, or cloud—handling quantization, ONNX export, tuning, and field monitoring under CPU and RAM constraints.
  • Work closely with data engineers and backend teams to ensure models run reliably within production pipelines, not just as prototypes.
  • Evaluate model success by their real-world impact, such as reducing unplanned downtime and improving output quality, beyond traditional metrics like F1 or AUC scores.
  • Contribute proactively to product roadmaps and technical decision-making, offering opinions and not merely executing assigned tasks.

Candidate Requirements

  • Bachelor’s degree or higher in Data Science, Computer Science, Mathematics, Physics, or related fields.
  • At least three years of hands-on experience with Python, popular machine learning frameworks such as PyTorch or scikit-learn, and deploying models in production environments.
  • Practical knowledge of time series analysis and signal processing, understanding limitations of naive neural networks on noisy industrial signals.
  • Basic familiarity with MLOps concepts including model and data versioning, reproducible pipelines, and ML code testing.
  • Strong English communication skills, both spoken and written.
  • Ability to justify technical decisions and represent them persuasively within a team, even against majority opinion, when well-founded.
  • Preferred: Experience with edge deployment technologies (ONNX, TensorRT, quantization), industrial protocols (OPC-UA, MQTT), or large language models (LLMs) for chat and autonomous tasks.

Benefits and Work Environment

  • Comprehensive responsibility spanning data acquisition design through pipeline development to model inference across edge, on-premises, or cloud infrastructure.
  • Autonomy in decisions regarding architecture, tooling, and test coverage within the team.
  • Use of advanced coding tools like Codex and Claude Code, exploring new development practices early.
  • Predominantly onsite work in Stuttgart, Ulm, or Berlin with projects in industrial IoT for customers globally, including mid-sized manufacturers and major corporations.

Additional Information

At DATATRONiQ, you will not be just a small cog in the machine but will actively shape products and contribute ideas. The role offers a unique chance to work on solutions with significant influence in industrial manufacturing. Candidates eager to tackle challenging problems and engage proactively are encouraged to apply. Please note that recruitment agencies and headhunters are asked not to contact.

Work styles they’re looking for

সিদ্ধান্ত গ্রহণ দলগত সহযোগিতা প্রযুক্তিগত যোগাযোগ Proactive Problem Solving

আপনি যদি উত্তর চান তবে এটি রেখে দিন — আমরা এটি অন্য কোনো কাজে ব্যবহার করব না।

ব্রাউজ করতে ক্লিক করুনড্র্যাগ অ্যান্ড ড্রপ, অথবা পেস্ট একটি স্ক্রিনশট

PNG, JPG, GIF, MP4, WebM, MOV · প্রতিটি সর্বোচ্চ ২০ মেগাবাইট · সর্বোচ্চ ৫টি ফাইল

🤖
অনলাইন · তাৎক্ষণিক এআই সহায়তা