Steve NyamiComputer Science & Medical AI

Computer science Medical AI

Intelligence.
With a human
perspective.

Hi, I’m Steve Nyami. A Computer Science master’s student at TU Darmstadt with experience in Medical AI.

Exploring the space between intelligent systems and better healthcare.

Explore my work A little about me
Computer vision in surgeryIllustrative
00:14:32:10

Frame capture. Surgical video runs at 25 frames a second or more, so a live system has at most 40 ms per frame. Each frame reaches the model as a grid of pixel values.

Example tools
  • FFmpeg
  • OpenCV

Feature extraction. A backbone network turns each patch of the frame into a feature vector, a list of numbers that describes what the patch shows.

Example models
  • ResNet
  • Vision Transformer
  • DINOv2

Segmentation. A segmentation model assigns every pixel a class, such as instrument, tissue, or background, which outlines each region.

Example models
  • U-Net
  • Mask R-CNN
  • SAM

Detection and tracking. A detector boxes each instrument, and a tracker links the boxes across frames, so each tool keeps its ID and its tip leaves a path.

Example models
  • YOLO
  • DETR
  • ByteTrack

Phase and action recognition. A temporal model combines recent frames to recognize the phase of the operation, such as dissection, and the action under way, such as retracting tissue.

Example models
  • TCN
  • LSTM
  • Transformer

Human review. A clinician confirms or corrects each result. Corrections become new training labels, so the models can keep improving.

Example methods
  • Human in the loop
  • Active learning

One frame in slow motion. A stylized scene, not real patient data or model output.

Curiosity. Context. Care.Scroll to discover

Currently studyingMaster’s in Computer Science

UniversityTU Darmstadt

Working studentFraunhofer IGD

01 Project portfolio

The work behind
the ideas.

A closer look at the problems, methods, and lessons behind my work.

Coming soon

Every project
has a story.

Project details will be added here, including the questions I explored, my contribution, and what I learned along the way.

  • Problem & context
  • Approach & methods
  • Results & reflections

More about my Medical AI experience and computer science work, coming soon.

02 Experience

Building, supporting,
and teaching.

Hands-on work in software engineering, IT operations, and mathematics teaching, alongside my studies at TU Darmstadt.

  1. – Present

    Darmstadt, Germany

    Current

    Working Student

    Fraunhofer IGD

    Developing and maintaining modern, scalable software components, and taking an active part in DevOps processes that automate, integrate, and stabilize production systems.

    • Software engineering
    • DevOps

  2. –

    Darmstadt, Germany

    Research Assistant

    Technische Universität Darmstadt

    Kept complex IT system landscapes in university operations secure and stable through efficient error analysis, system administration, and network management.

    • System administration
    • Networking
    • Error analysis

  3. –

    Darmstadt, Germany

    Mathematics Tutor

    Technische Universität Darmstadt

    • Led exercise sessions on proof principles, linear algebra, systems of linear equations, vector spaces, and linear maps.
    • Corrected student submissions.
    • Met regularly with the course’s teaching staff.
    • Linear algebra
    • Teaching

Recognition

  • Freudenberg Award — Digital Science2026

Certifications

  • Data Analysis in Python: Using Pandas DataFrames
  • Working with BigQuery
  • COVID-19 Data Analysis Using Python
  • Introduction to Data Analysis Using Microsoft Excel

Top skills

  • Python
  • Software development
  • Surgical data science
  • Data analysis

03 Education

Grounded in
computer science.

Both degrees at Technische Universität Darmstadt: a bachelor’s in Informatik, followed by a master’s in Computer Science.

  1. Master’s degree

    M.Sc. Computer Science

    Technische Universität Darmstadt

    Current
  2. Bachelor’s degree

    B.Sc. Informatik

    Technische Universität Darmstadt

04 Guiding principles

Good questions.
Thoughtful progress.

A perspective on building useful technology.

  1. 01

    Start with context

    Understand the problem, the people, and the setting before choosing a solution.

  2. 02

    Explore with curiosity

    Connect ideas, ask better questions, and make room for learning.

  3. 03

    Make it understandable

    Value clear explanations and keep the reasoning behind a result visible.

  4. 04

    Reflect and improve

    Consider the evidence, acknowledge limitations, and build on what was learned.

05 A little about me

Hi, I’m Steve.

A student of computer science.
Curious about its human possibilities.

I’m pursuing a master’s in Computer Science at TU Darmstadt and have experience in Medical AI. Alongside my studies, I’m a working student at Fraunhofer IGD, where I build and maintain software components. My interests sit where technical ideas meet the complexity of healthcare, computer vision in particular.

This is a space to share my work, the thinking behind it, and what I learn along the way.

Let’s connect

06 Connect

A good conversation
can be a beginning.

Interested in computer science, Medical AI, or the ideas that connect them? There’s always more to explore together.

Let’s stay in touch.

Send me an email or connect with me on LinkedIn.