Über Pegah
Applied AI & Automation Specialist
I design and implement intelligent systems that automate decision-making, reporting, and communication.
My focus is on secure, private, and high-impact AI integrations that blend seamlessly into existing workflows, from GPT-powered data assistants to document summarization and business process automation.
- Workflow automation (Python, LangChain, GPT-4, APIs
- AI assistants for business data and reports
- Chat + document bots with secure backends
- Prompt and system design for reliability and context awareness
Full-Stack Engineer
I build end-to-end applications — from concept to production — that connect data, logic, and user experience.
My strength is creating robust backends with clean, intuitive UIs, ensuring that the intelligence behind the system is actually usable.
- Frontend: React, TypeScript, Next.js, Tailwind
- Backend: Rust, Python (FastAPI, Flask), Node.js
- APIs, authentication, and deployment (Docker, Kubernetes, Azure)
- Real-time and event-driven app design
Data Engineer
I design and manage data platforms that scale securely — pipelines, warehouses, and transformations that feed analytics and AI models.
I focus on clean architecture, automation, and observability.
- Data pipelines: Airflow, dbt, Kafka, Azure Data Factory
- Cloud: Azure, AWS, Kubernetes
- Databases: MySQL, Postgres, Snowflake
- Monitoring, alerting, and data quality automation
Englisch
Muttersprachlich oder zweisprachig
Deutsch
Konversationssicher
Persisch
Muttersprachlich oder zweisprachig
Vor Ort möglich
Cologne (bis zu 50 km)
Projekt- und Berufserfahrung
- BASF Digital Farming,Senior Scientist and EngineerDezember 2021 - Dezember 2023 (2 Jahre)
- Deployed ML models to the cloud to support growth stage modelling.
- Automated the transformation of binary field data into decision-ready agronomic maps.
- Developed Airflow DAGs to clean and enrich geospatial and weather time series for
- precision agriculture.
- Built scalable weather data pipelines using chunked, compressed N-dimensional Zarr arrays for efficient cloud storage and analytics.
- Capgemini,Senior Data EngineerDezember 2023 - Heute (2 Jahre und 6 Monate)
- I returned back to support one of my first clients since starting at Capgemini, this time focusing on the migration of their Cloudera Data Engineering workflows to AWS. Our approach involves rearchitecting key ETL processes to run on AWS Lambda and Python based libraries for greater scalability and cost efficiency. Transformed and loaded data is stored as Parquet files, with Impala tables enabling users to run queries on top of this structured data.
- As a Work Stream Lead, I spearheaded the migration of complex, non-vendor applications from on-premises infrastructure to an Azure tenant, operating under tight deadlines. This role required a diverse skill set encompassing Azure cloud architecture, infrastructure management, Node.js development, project management, and business analysis;
- We replaced a costly legacy system for processing interbank exchange rates with a more efficient and affordable cloud-based solution. The new system consolidates data from various sources and distributes it to multiple stakeholders. Leveraging AWS cloud technologies like Spark, Glue, Step Functions, and Lambda, we developed data pipelines and the associated Infrastructure as Code using Terraform. This modernized system reduced operational costs, streamlined data management, and improved overall efficiency.
- MetService,Research ScientistFebruar 2020 - Dezember 2021 (1 Jahr und 10 Monate)My main driver at MetService was modernizing the data landscape and removing typical obstacles for data scientists and product developers.Notable contributions include:Prototyping:
- ConvLSTM for spatial Time Series forecast with potential applications in precipitation nowcasting and wind ramp events and fog forecast;
- A small application for streaming meteorological variables using Apache Kafka (event streaming), Confluent ksqlDB (streaming ETL), Graphite (Time Series DB) and Grafana (dashboarding/visualization);
Development:- Built and deployed deep learning models (ConvLSTM) for spatial time series forecasting, applied to precipitation nowcasting and fog prediction.
- Created streaming architecture prototypes with Kafka, ksqlDB, and Grafana to enable real time weather data visualization.
- Developed cloud-native ML solutions for weather prediction using containers (ECS) and serverless components (AWS Lambda).
Empfehlungen
Sei die erste Person, die Pegah empfiehlt
Teile Deine Erfahrung aus der Zusammenarbeit mit diesem Freelancer.
Diese Freelancer passen auch zu Ihren Kriterien
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Ausbildung und Abschlüsse
- Doctor of Philosophy - PhDVictoria University of Wellington2016Doctor of Philosophy - PhD
- Master of Engineering - MEngUniversity of TehranMaster of Engineering - MEng