Jan Krepl, Developer in Geneva, Switzerland
Jan is available for hire
Hire Jan

Jan Krepl

Verified Expert  in Engineering

Machine Learning Engineer and Developer

Location
Geneva, Switzerland
Toptal Member Since
July 14, 2023

Jan是一位对软件工程充满热情的机器学习工程师, machine learning, leadership, and online education. 他在应用计算机视觉方面有丰富的专业经验, natural language processing, 以及在学术和商业环境中的时间序列分析. 他还将大部分空闲时间用于开源软件和教育内容创作.

AlgorithmsMachine LearningNatural Language Processing (NLP)Artificial Intelligence (AI)Back-end DevelopmentComputer VisionBack-endLarge Language Models (LLMs)ServerlessOpenAIPineconeChatGPTOpenAI GPT-4 APIPythonScikit-learnBERTRetrieval-Augmented Generation (RAG)StatsModelsGensimSphinxVertex AI

Portfolio

The EPFL
Python, CI/CD pipeline, BERT, PyTorch, Elasticsearch...
The EPFL
CI/CD管道，Python, OpenCV, Keras, PyTorch，深度学习...
Nectar Financial
Python，机器学习，深度学习，Gensim, Scikit-learn, Pandas, NumPy...

Experience

Python - 9 yearsMachine Learning - 7 yearsScikit-learn - 7 yearsNatural Language Processing (NLP) - 7 yearsDeep Learning - 6 years机器学习操作(MLOps) - 5年Computer Vision - 5 yearsPyTorch - 5 years

Availability

Part-time

Preferred Environment

Python, Machine Learning, Notion

The most amazing...

...我开发了一个从科学论文中提取知识的问答工具.

Work Experience

机器学习部门经理|蓝脑项目

2022 - PRESENT
The EPFL
  • 设计了一个以语义检索为核心的文献检索系统, question answering, named entity recognition, and entity linking, built on top of recent large language models. 整个系统通过Kubernetes和AWS大规模部署.
  • 管理一个由四名经验丰富的机器学习工程师组成的团队.
  • 作为执行最佳实践的首席开发人员.
Technologies: Python, CI/CD pipeline, BERT, PyTorch, Elasticsearch, Natural Language Processing (NLP), Amazon Web Services (AWS), Machine Learning Operations (MLOps), Deep Learning, Machine Learning, Docker, Unit Testing, REST APIs, FastAPI, GitLab, GitHub, NumPy, Kubernetes, Vim Text Editor, Shell Scripting, SQL, SpaCy, Agile Software Development, Data Science, Orchestration, Computer Vision, Data Versioning, Apache Airflow, LangChain, Sphinx, TensorBoard, Hugging Face Transformers, Git, Generative Pre-trained Transformers (GPT), ChatGPT, OpenAI GPT-4 API, Artificial Intelligence (AI), Hugging Face, APIs, JavaScript, Regular Expressions, Product Consultant, Natural Language Understanding (NLU), GPT, Algorithms, Back-end Development, Language Models, Amazon S3 (AWS S3), Amazon EC2, AWS Glue, Terraform, Pytest, GitLab CI/CD, Large Language Models (LLMs), Redis, Redis Cache, Technical Leadership, Leadership, Retrieval-augmented Generation (RAG), OpenAI, NoSQL, Back-end, Asyncio, Containerization, Python Asyncio, Test-driven Development (TDD), AWS Lambda, REST, Serverless, SDKs, Pinecone

机器学习工程师|蓝脑计划

2018 - 2022
The EPFL
  • 构思并实现了一种2D脑切片图像配准的监督算法，该算法成为内部工作流程的一部分.
  • 开发了一个科学文章的知识抽取管道，主要功能有解析等, neural search, and named entity recognition.
  • 直接参与各种神经科学项目, 包括用图神经网络进行神经元类型分类和用生成对抗网络进行形态学图像合成.
Technologies: CI/CD管道，Python, OpenCV, Keras, PyTorch，深度学习, Machine Learning, GitLab CI/CD, Image Registration, SpaCy, Git, Machine Learning Operations (MLOps), REST APIs, FastAPI, GitLab, GitHub, NumPy, Kubernetes, Vim Text Editor, Shell Scripting, SQL, Agile Software Development, Data Science, Orchestration, Natural Language Processing (NLP), Elasticsearch, Docker, Computer Vision, Data Versioning, Apache Airflow, Unit Testing, Sphinx, TensorBoard, Hugging Face Transformers, MySQL, PostgreSQL, Artificial Intelligence (AI), Hugging Face, APIs, Regular Expressions, Natural Language Understanding (NLU), GPT, Algorithms, Back-end Development, Language Models, TensorFlow, Pytest, Large Language Models (LLMs), NoSQL, Back-end, Asyncio, Containerization, Python Asyncio, Test-driven Development (TDD)

Data Scientist

2018 - 2018
Nectar Financial
  • 增强内部投资组合优化算法与回报预测使用监督学习技术. 增加了自定义约束和目标函数，使工具更加灵活.
  • 应用文本嵌入算法，如Doc2Vec和TF-IDF，在对冲基金的情况说明书和报告. 反过来，这些嵌入被用于聚类，这允许更好的多样化.
  • 考虑到各种对冲基金特定的限制，如锁定期，开发了一个定制的后台测试框架.
Technologies: Python，机器学习，深度学习，Gensim, Scikit-learn, Pandas, NumPy, Jupyter Notebook, SciPy, StatsModels, SpaCy, REST APIs, GitHub, SQL, Data Science, Natural Language Processing (NLP), Keras, Docker, Time Series Analysis, Unit Testing, Git, Artificial Intelligence (AI), JavaScript, Regular Expressions, Algorithms, Back-end Development, Pytest, NoSQL, Test-driven Development (TDD)

Quantitative Risk Analyst

2016 - 2017
UBS
  • 使用Visual Basic维护Lombard贷款部门的压力测试代码库, SQL, and SAS.
  • 定期生成风险报告，作为其他部门的输入.
  • 支持高级分析师创建自定义风险模型.
Technologies: SQL, SAS, Excel VBA, Shell Scripting, Probability Theory, Time Series Analysis, Algorithms, NoSQL

Mildlyoverfitted | Educational Videos

http://www.youtube.com/@mildlyoverfitted/
我开发的一个YouTube频道，用来存放教育内容和资源. 这个频道的特色是我制作的关于机器学习、深度学习和Python的视频. 其中一个主要目标是解释事物是如何在底层工作的，以及我们如何从头开始实现解决方案.

DeepDow |深度学习组合优化

http://github.com/jankrepl/deepdow/
一个Python包，用于深度学习的投资组合优化. 它试图合并以下两个非常常见的投资组合优化步骤:
• Forecasting the market's future evolution, 如长短期记忆网络(LSTM)和广义自回归条件异方差(GARCH)。.
•提供优化问题的设计和解决方案，如凸优化.

它通过构造一个层的管道来实现. 最后一层执行分配，前面的所有层都作为特征提取器. The overall network is fully differentiable, 并且可以通过梯度下降算法对其参数进行优化.

MLtype | Command Line Tool

http://github.com/jankrepl/mltype/
一个程序员友好的命令行工具，用于提高输入速度和准确性. 主要目标是帮助程序员练习编程语言. It uses neural networks to generate text. 人们可以使用预先训练好的网络，也可以从头开始训练新的网络.

Atlas对准|多模态配准与对准

http://github.com/BlueBrain/atlas-alignment/
执行多模态图像配准的工具箱, 其中包括传统的和监督的深度学习模型. 该项目起源于蓝脑计划，旨在通过原位杂交(ISH)基因表达和尼氏染色获得小鼠脑图谱. 这个项目发表在前沿媒体上，你可以通过这个链接访问:frontiersin.org/articles/10.3389/fninf.2021.691918/full/

PyChubby | Automated Face-warping Tool

http://github.com/jankrepl/pychubby/
A Python package for automated face warping. 它允许用户通过编程改变图像中任何人的面部表情和形状. 它是基于使用计算机视觉的几何变换.

Languages

Python, SQL, SAS, Excel VBA, JavaScript

Libraries/APIs

PyTorch, Scikit-learn, NumPy, Keras, SciPy, Pandas, Matplotlib, REST APIs, TensorFlow, Asyncio, Python Asyncio, JAX, OpenCV, SpaCy, React

Tools

Vim Text Editor, Git, GitLab CI/CD, Pytest, TensorBoard, GitLab, GitHub, Notion, Amazon SageMaker, Cloud Dataflow, Google Compute Engine (GCE), AWS Glue, Terraform, Inkscape, Apache Airflow, Adobe Premiere Pro, Seaborn, Gensim, StatsModels, Scikit-image, Google Kubernetes Engine (GKE)

Paradigms

单元测试、数据科学、测试驱动开发(TDD)、REST、Scrum、敏捷软件开发

Platforms

Kubernetes, Docker, Amazon Web Services (AWS), Jupyter Notebook, Vertex AI, Amazon EC2, Google Cloud Platform (GCP), AWS Lambda

Storage

Elasticsearch, Google Cloud Storage, Amazon S3 (AWS S3), Redis, Redis Cache, NoSQL, Neo4j, MySQL, PostgreSQL

Other

Probability Theory, Mathematical Analysis, Linear Algebra, Statistics, Machine Learning, Portfolio Optimization, Orchestration, Machine Learning Operations (MLOps), Shell Scripting, Generative Pre-trained Transformers (GPT), BERT, Hugging Face Transformers, Sphinx, Natural Language Processing (NLP), Finance, Computer Vision, ChatGPT, OpenAI GPT-4 API, Artificial Intelligence (AI), Hugging Face, APIs, Regular Expressions, Natural Language Understanding (NLU), GPT, Algorithms, Back-end Development, Language Models, Pub/Sub, Large Language Models (LLMs), Technical Leadership, Leadership, Retrieval-augmented Generation (RAG), OpenAI, Back-end, Containerization, Serverless, SDKs, Pinecone, Optimization, Microeconomics, Macroeconomics, Mathematical Finance, Quantitative Risk Analysis, Numerical Methods, MLflow, FastAPI, LangChain, Time Series Analysis, Product Consultant, Measure Theory, Econometrics, Private Company Valuation, Deep Learning, Scrum Master, CI/CD Pipelines, Online Course Design, Recurrent Neural Networks (RNNs), Open Source, Image Registration, Data Versioning, Google BigQuery, Full-stack Development

Frameworks

Apache Spark

2015 - 2018

Master's Degree in Quantitative Finance

ETH Zurich - Zurich, Switzerland

2011 - 2014

Bachelor's Degree in Economics

Charles University - Prague, Czechia

FEBRUARY 2024 - FEBRUARY 2026

HashiCorp认证:Terraform Associate (003)

HashiCorp

JANUARY 2024 - JANUARY 2027

AWS Certified Solutions Architect - Associate

Amazon Web Services

SEPTEMBER 2023 - SEPTEMBER 2025

谷歌云认证专业机器学习工程师

Google Cloud

AUGUST 2023 - AUGUST 2026

AWS Certified Machine Learning - Specialty

Amazon Web Services

JUNE 2023 - JUNE 2025

Databricks认证的Apache Spark 3助理开发人员.0

Databricks Inc.

MARCH 2023 - MARCH 2026

AWS Certified Cloud Practitioner

Amazon Web Services

FEBRUARY 2023 - FEBRUARY 2026

CKAD:认证Kubernetes应用程序开发人员

The Linux Foundation

JANUARY 2023 - PRESENT

Professional Scrum Master (PSM I)

Scrum.org

JULY 2015 - PRESENT

CFA Level I (Passed)

CFA Institute

Collaboration That Works

How to Work with Toptal

在数小时内，而不是数周或数月，我们的网络将为您直接匹配全球行业专家.

1

Share your needs

在与Toptal领域专家的电话中讨论您的需求并细化您的范围.
2

Choose your talent

在24小时内获得专业匹配人才的简短列表，以进行审查，面试和选择.
3

Start your risk-free talent trial

与你选择的人才一起工作，试用最多两周. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring