Mika Senghaas Avatar
Hi 👋🏻, I'm Mika. I am a master student at EPFL and research engineer at Prime Intellect.

Open Source

Research

2025

SYNTHETIC-2: Scaling Distributed Synthetic Data Generation for Verified Reasoning

M. Senghaas, J. Ong, M. Basra, J. Mattern, J. Straube, S. Jaghouar, J. Hagemann
BlogThreadHugging Face

INTELLECT-2: A Reasoning Model Trained Through Globally Decentralized Reinforcement Learning

S. Jaghouar, J. Mattern, J. Ong, J. Straube, M. Basra, A. Pazdera, K. Thaman, M. Di Ferrante, F. Gabriel, F. Obeid, K. Erdem, M. Keiblinger, M. Senghaas, J. Hagemann
ReportBlogHugging Face

DiLoCo-SWARM: Towards Scalable Decentralized Training Across the Globe

M. Senghaas, M. de Vos, R. Sharma, A. Dhasade
ReportSlidesGitHubW&B

2024

CNNs for Style Transfer of Digital to Film Photography

M. Senghaas, A. O'Toole, P. Lardet
ReportGitHub

Finetuning and Quantizing Phi-3 Mini for MCQA

M. Senghaas, L. Cizinsky, P. Lardet
ReportGitHubHF

2023

Advancing Homepage2Vec with LLM-Generated Datasets for Multilingual Website Classification

M. Senghaas, L. Cizinsky, P. Nutter
ReportDatasetGitHubDemo

Benchmarking SOT Feature Transform for Biomedical Few-Shot Learning Tasks

M. Senghaas, A. Barla, L. Cizinsky
ReportGitHub

Deep Learning Methods for Mutagenicity Prediction in Chemical Compounds

M. Senghaas
ReportGitHub

Navigating Indoors with Computer Vision - Deep Learning Approaches for Room-Level Indoor Localisation

M. Senghaas, S. Grasshof
ReportGitHub

Projects & Initiatives

Title Image for AddisCoder

AddisCoder

August 2024

Title Image for MegaBites

MegaBites

February 2024

Title Image for AITU

AITU

February 2023

Reach me

© Mika Senghaas 2025