Hello, I'm Yafi.

Computational Biology Researcher & Software Engineer

Yafi's Profile Photo

About Me

I am an aspiring PhD student for the Fall 2026 intake. My research interests lie at the intersection of AI and Computational Biology. I have developed a novel machine learning algorithm to design gene panels for spatial transcriptomics. With + years of software development experience, I currently lead an Agentic AI team at IQVIA, where we leverage foundational models to create intelligent assistants.

Education

Bachelor of Science in Computer Science & Engineering

Bangladesh University of Engineering and Technology

February 2017 - May 2022

CGPA: 3.73/4.00

Academic Transcript: View Transcript (Google Drive)

Notable Courses: Bioinformatics, Machine Learning, Operating System, Software Engineering, Compiler, Computer Security, Database, Computer Networks, Computer Architecture, Artificial Intelligence, Microprocessors and Microcontrollers, Data Structures and Algorithms, Object Oriented Programming Language.

Test Scores

Research & Publications

Research Interest

My goal is to bridge the gap between computer science and life sciences by developing accessible, high-impact tools. My research interests lie in Computational Biology, with a focus on single-cell RNA sequencing, spatial transcriptomics and high-dimensional data visualization. I am particularly interested in building algorithms that can:

  • Solve complex biological mysteries through multi-omics integration.
  • Generate valuable insights for medical diagnosis and treatment.
  • Empower researchers and clinicians with interpretable software.

Computational Biology Bioinformatics Single-Cell Technologies Machine Learning Artificial Intelligence Large Language Models

scGIST: Gene Panel Design for Spatial Transcriptomics with Prioritized Gene Sets

Published in Genome Biology on February 2024

The initial goal of the project was to modify a simple linear programming algorithm to include prioritized genes which we achieved in a short period of time. After that, we challenged ourselved to achieve even better results. Through several iterations and experiments, we developed scGIST.

  • Developed a deep-learning tool for designing single-cell spatial transcriptomics (sc-ST) gene panels.
  • Created a custom loss function for constrained feature selection, optimizing both cell type detection and inclusion of prioritized genes.
  • Achieved high accuracy in cell type identification across multiple datasets, outperforming existing methods, especially for small panel sizes.
  • Enabled researchers to systematically prioritize genes of interest, including easy-to-target genes and receptor-ligand complexes, without sacrificing accuracy.

Publications

  • Yafi, M., Hisham, M., Grisanti, F. et al. scGIST: gene panel design for spatial transcriptomics with prioritized gene sets. Genome Biol 25, 57 (2024). https://doi.org/10.1186/s13059-024-03185-y

Experience

Software Development Engineer II

IQVIA

October 2024 - Present

  • Team lead of an AI Assistant team, overseeing design, development, and deployment process.
  • Implement agentic AI solutions using LangGraph and LangChain, focusing on modularity and scalability.
  • Redesigned agent graph to reduce latency by 15% and token usage by 10%.
  • Monitor and troubleshoot issues in LangSmith to maintain reliability.
  • Build and deploy agent server in Kubernetes hosted in AWS.
  • Winner of the IQVIA Platinum Award for managing major change requests in the Agent Server.

Software Development Engineer I

IQVIA

September 2023 - September 2024

  • Led .NET backend development team.
  • Designed and implemented scalable microservices and API-driven solutions.
  • Optimized the PowerPoint export feature, reducing memory usage by 85%.
  • Delivered performance improvements of APIs with response time gains of up to 10x.
  • Resolved production issues with SREs and maintained dev Kubernetes cluster hosted in AWS.
  • Received the IQVIA Gold Award for performance optimization.

Software Engineer

Enosis Solutions

May 2022 - August 2023

  • Developed a web application using .NET and Angular, following the Kanban methodology.
  • Automated performance metrics retrieval from the Google Ad Manager API using timer-triggered Azure Functions.
  • Optimized complex queries in Microsoft SQL Server, significantly improving data retrieval performance.
  • Created automated statistics reports and scheduled email delivery using timer-triggered Azure functions.
  • Migrated legacy .NET MVC project to .NET Core.

Projects

Aero Color

  • A machine learning model to colorize any given grayscale image.
  • Created a U-Net model using Keras to train on 25,000 images.
  • Used OpenCV to pre-process and post-process the images.
Python Keras Tensorflow

Aero Code

  • An online website to upload and review codes, create groups, and send personal messages.
  • Used ReactJS, NextJS, and MongoDB to develop the frontend and backend of the website.
  • Used docker to deploy the website.
NextJS React Node

Aero Mouse

  • A wireless 3D motion-controlled mouse built using an ATmega32 microcontroller.
  • Used Atmel Studio to write the firmware for the mouse in C.
  • Created a driver program in Java to make a serial connection with the computer.
C Java ATmega32

Skills

Languages

Python C# JavaScript Java C++ R

Frameworks

LangGraph Keras Tensorflow Seurat .Net Angular React

Tools

Git Docker AWS Azure Kubernetes

Resume

Academic CV

View CV

Get In Touch

Let's connect and discuss opportunities