Skip to content
View farabi1038's full-sized avatar

Highlights

  • Pro

Block or report farabi1038

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
farabi1038/README.md

👋 Hi there! I'm Farabi!

📊 Applied Scientist at Amazon | 🎓 Ph.D. in Computer Science | 🔬 Applied AI Researcher

Ph.D., Computer Science — Iowa State University • Oct 2025
📍 Current Role: Applied Scientist at Amazon, California


"Bridging cutting-edge AI research with real-world impact across transportation, agriculture, and quantum computing"


Website LinkedIn Google Scholar GitHub Email


📄 Download Industry ResumeDownload Academic Resume


🎯 About Me

I'm an AI researcher and applied scientist with expertise spanning multiple high-impact domains. My work bridges academic research and practical industry applications, with a focus on deploying AI solutions that create measurable real-world impact.

🔬 Research Impact

  • 📚 16+ Publications at top-tier venues
  • 🏆 13 Papers Accepted in 2026 to ICML, ACL, CVPR, ITSC, and IEEE T-ITS
  • 🎯 10+ Under Review / In Submission at NeurIPS, ECCV, ACL ARR, and IEEE T-ITS
  • 🧑‍⚖️ Area Chair for ACL 2026 Industry Track
  • 👥 Reviewer for NeurIPS, ICML, CVPR, ICCV, ICLR, ACM MM, WACV

💼 Industry Impact

  • 🏢 Amazon Applied Scientist II (January 2026 - Present)
  • 🏢 Amazon Applied Scientist Intern (Summer 2025)
  • 🌾 35% Accuracy Boost in soil mapping
  • 🚗 Iowa DOT Partnership for safety systems
  • 💰 20% Cost Reduction via cloud optimization

🌟 Domain Expertise

🤖 Large Language Models & Generative AI

Pioneering multimodal AI for safety-critical applications

🎥 Video-LLMs for Safety

  • Crash detection systems
  • Automated narrative generation
  • Real-time video analysis
  • Synthetic data simulation
  • Published at TRC-30 '24

⚡ Model Efficiency

  • Mamba state-space pruning
  • Edge deployment optimization
  • Cache-efficient sampling
  • Resource-constrained AI
  • 2 Papers at EMNLP '25
  • Compression work accepted at ICML '26 & ACL '26

🧠 LLM Theory & Reasoning

  • Counterfactual sensitivity
  • Risk & conformal certification
  • Reward function analysis
  • Faithful reasoning
  • 4 Papers at ICML '26, 4 at ACL '26 (accepted)

Key Publications:

  • ICML 2026 - CalPro: Prior-Aware Evidential Conformal Prediction for Protein Structures
  • ACL 2026 - Causal Consistency Regularization for Verifiably Sensitive Reasoning (Main)
  • ACL 2026 - Beyond Variance: Knowledge-Aware LLM Compression via Fisher-Aligned Subspace Diagnostics (Main)
  • EMNLP 2025 - Efficient Unstructured Pruning of Mamba State-Space Models
  • EMNLP 2025 - Cache-Efficient Posterior Sampling for RL with LLM-Derived Priors


🚗 Transportation & Autonomous Systems

AI-driven solutions for safer roads and smarter mobility

🎬 Computer Vision & Video Analytics

  • Real-time crash detection with temporal localization
  • HybridMamba for fine-grained action recognition
  • All-weather image classification (CycleGAN + SigLIP-2)
  • Change-point detection for action localization (CVPR '24)
  • Ensemble learning for sidewalk detection (ITSC '23)
  • LLM-Guided document layout analysis (CVPR '26)

🎯 Impact Metrics:

🤖 Reinforcement Learning & Control

  • Autonomous vehicle decision-making algorithms
  • LLM-guided meta-prior RL for traffic signal synchronization (ITSC '26)
  • Reward hacking detection & mitigation
  • Multi-agent coordination systems

🛠️ Simulation Frameworks:

  • 🚙 CARLA - Autonomous driving simulation
  • 🛣️ SUMO - Traffic flow modeling
  • 🏋️ Isaac Gym - Physics-based RL
  • 🎮 OpenAI Gym - Custom environments

🏢 Industry Projects:

  • ❄️ Iowa DOT Snowplow Navigation (ADAS)
  • 🚨 Real-time Incident Detection Systems
  • 🎥 Synthetic Crash Video Generation


🌾 Digital Agriculture & Environmental Analytics

Precision farming through geospatial ML and environmental modeling

🗺️ Precision Agriculture & Geospatial AI

  • Soil Mapping Engine: Production system serving 10+ farms
  • Performance: 35% accuracy improvement over baseline
  • Cost Optimization: 20% speed increase with reduced hosting costs
  • Cloud Integration: QGIS scripts for seamless AWS processing
  • Stack: Python • QGIS • ArcGIS • AWS • PostgreSQL

🌍 Key Capabilities:

  • Digital mapping & remote sensing
  • Terrain modeling & spatial analytics
  • Multi-source geospatial data fusion
  • Real-time environmental monitoring

🌱 Environmental Data Science

  • Climate data analysis & sustainability metrics
  • Resource allocation optimization
  • Crop yield prediction modeling
  • ML-driven agricultural decision support

📊 Technical Stack:

Geospatial:  QGIS • ArcGIS • GDAL
Cloud:       AWS (S3, Lambda, Athena)
Processing:  Spark • PySpark • PostGIS
ML:          Scikit-learn • XGBoost
Viz:         Tableau • Plotly • Kepler.gl

💼 Industry Experience:

  • 🏢 Data Scientist at SoilSerdem (2024)
  • 🌾 Led cross-departmental ML initiatives
  • 📈 Deployed production-grade prediction models


🔬 Advanced AI & Quantum Computing

Exploring the frontiers of machine learning and quantum algorithms

⚛️ Quantum Neural Networks

  • Novel QNN architectures
  • 10% boost in anomaly detection
  • Zero-trust security frameworks
  • Quantum algorithm design

🛠️ Frameworks:

  • torchquantum
  • Qiskit
  • PennyLane
  • StrawberryFields

📄 Publications:

  • Measurement: Digitalization '25
  • ECAI '25 (Quantum Spin Systems)

🧬 Biomedical AI Applications

  • RL-guided CAR T-cell optimization
  • Cellular behavior modeling
  • Therapy parameter tuning
  • Bio-monitoring with deep learning

📊 Research Impact:

  • Published in Biotechnology and Bioengineering
  • Heart/respiratory rate prediction
  • Enzyme classification models

🔍 Theoretical Foundations

  • Differentiable entropy regularization
  • Self-supervised few-shot learning
  • Conformal & risk certification
  • Information-lift statistics
  • Subspace-manifold methods (Grassmannian/Stiefel)

📝 In Submission:

  • NeurIPS '26 (5 submissions)
  • ACL ARR May '26 (6 submissions)
  • Novel theoretical frameworks

🚀 Featured Projects

🌟 GenAI-Powered Applications

GenAI-powered language learning platform with interactive exercises

Tech Stack: LangChain OpenAI Streamlit Python

Features:

  • 💬 Conversational AI tutor
  • 📝 Grammar correction
  • 🎯 Personalized learning paths
  • 🔊 Pronunciation feedback

Complete PyTorch implementation of Google's PaliGemma multimodal model

Tech Stack: PyTorch Transformers Computer Vision

Features:

  • 🖼️ Vision-language understanding
  • 🏗️ Built from architectural principles
  • 📚 Educational annotations
  • ⚡ Optimized inference

💻 Technical Implementations & Research

Project Description Tech Stack Links
🔍 Wiki Word Analyzer Web-based text analysis with intelligent caching FastAPI Redis NLP Code
🚦 RL Signal Sync Reinforcement learning for traffic signal optimization Ray RLlib SUMO Code
🏆 Mcity AV Challenge Autonomous vehicle competition entry Python Computer Vision Code
📚 Online Course Projects Collection of ML/AI implementations PyTorch TensorFlow Code


📖 Academic Coursework

Comprehensive graduate-level training in AI, ML, and computer science

🤖 Machine Learning & AI

Course Code Course Title Repository
COMS 573 Machine Learning 📂 View
COMS 574 Intro to Machine Learning 📂 View
COMS 575 Computational Perception & CV 📂 View
COMS 572 Artificial Intelligence 📂 View

🛠️ Systems & Applications

Course Code Course Title Repository
COMS 576 Motion Strategy Algorithms 📂 View
ME 592 Data Analytics & ML for CPS 📂 View

💼 Professional Experience

🎓 Academia

Lecturer
Dallas College
Aug 2025 - Jan 2026


Graduate Research Assistant
Iowa State University
Jan 2022 - May 2025


Graduate Teaching Assistant
Iowa State University
Jan 2024 - May 2025 (COM S 227, 576)


Graduate Teaching Assistant
University of Vermont
Aug 2019 - May 2020

🏢 Industry

Applied Scientist II
Amazon Inc.
Jan 2026 - Present


Applied Scientist Intern
Amazon Inc.
Jun 2025 - Aug 2025


Data Scientist
SoilSerdem
Jan 2024 - Dec 2024


Data Engineer Intern
Etalyc Inc.
May 2021 - Jul 2021

🔬 Research Focus

LLMs & Multimodal AI
Video understanding
Model efficiency


Transportation Safety
Crash detection
Autonomous systems


Precision Agriculture
Geospatial ML
Environmental analytics


Quantum Computing
QNN architectures
Security applications


📝 Selected Publications

🏆 Recent Acceptances (2026)

✅ ICML 2026 (4 Papers)

  • CalPro: Prior-Aware Evidential Conformal Prediction for Protein Structures
  • What Reward Structure Enables Efficient Sparse-Reward RL?
  • CGRiC: Compositional Risk Certification for Structured LLM Outputs
  • Certificate-Guided Pruning for Stochastic Lipschitz Optimization

✅ ACL 2026 (4 Papers — 3 Main, 1 Findings)

  • Adaptive Constraint Propagation (Main)
  • Beyond Variance: Knowledge-Aware LLM Compression (Main)
  • Causal Consistency Regularization (Main)
  • Detecting Proxy Gaming in RL and LLM Alignment (Findings)

✅ IEEE/CVF CVPR 2026

  • LLM-Guided Probabilistic Fusion for Label-Efficient Document Layout Analysis

✅ IEEE ITSC 2026 (2 Papers)

  • Beyond Accuracy: Empirical Spectral Diagnostics for Crash Classification Models
  • LLM-Guided Meta-Prior RL for Adaptive Traffic Signal Synchronization

✅ Journal Publications (2026)

  • Image Segmentation with LLMs: Survey - IEEE T-ITS (IF 8.4)
  • Improving Crash Data Quality via NLP - Journal of Safety Research (IF 4.8)

🏆 Earlier Acceptances (2023–2025)

✅ ECAI 2025

  • HMAE: Self-Supervised Few-Shot Learning for Quantum Spin Systems

✅ EMNLP 2025 (2 Papers)

  • Efficient Unstructured Pruning of Mamba State-Space Models
  • Cache-Efficient Posterior Sampling for RL with LLM-Derived Priors

✅ ITSC 2025 (2 Papers)

  • ClearVision: All-Weather Classification in Traffic Imagery
  • Expert-Aligned Evaluation of Crash Narrative Models

✅ Journal Publications

  • Quantum-driven Zero Trust Architecture - Measurement: Digitalization (2025)
  • RL-Guided CAR T-Cell Control - Biotechnology and Bioengineering (2024)

✅ IEEE/CVF CVPR 2024

  • DeepLocalization: Change Point Detection for Temporal Action Localization

✅ IEEE ITSC 2023

  • Robust Sidewalk Detection with Ensemble Learning

📝 In Submission (2026)

NeurIPS 2026 (5 Submissions)

  • Learning to Forget Attention (CRAM)
  • On the Sparsifiability of Correlation Clustering
  • Grassmannian Mixture-of-Experts
  • Auditing Near-Optimal Policies
  • Topology-Aware State Abstraction

ACL ARR May 2026 (6 Submissions)

  • Calibrated Adaptation: Bayesian Stiefel Manifold Priors
  • Anytime-Valid Answer Sufficiency Certificates
  • EST-PRM: Stress-Testing Process Reward Models
  • Grounded Decoding for Faithful RAG
  • Dynamic Proxy-Mixing
  • Canonicalized Stable-List Replay

ECCV 2026 & IEEE T-ITS

  • HybridMamba for Sub-Second Temporal Localization
  • Temporal Zoom Networks
  • Calibrated & Resource-Aware Super-Resolution

🧑‍⚖️ Academic Service & Leadership

🎯 Area Chair & Peer Review for Premier Venues

🎥 Computer Vision CVPR '24–'26
ICCV '25
WACV '26–'27

🧠 Machine Learning NeurIPS '26
ICML '26
ICLR '26
ECAI '25

💬 NLP ACL Industry Track '26 (Area Chair)
ACL ARR / EMNLP '26

🛣️ Transportation / Multimedia ITSC '23–'25
ACM MM '25–'26
IEEE ICVES '25

📊 Total Reviews: 30+ papers across 10+ top-tier venues • 🧑‍⚖️ Area Chair: ACL 2026 Industry Track

🌟 Leadership & Mentoring

🏛️ University Service

  • 👥 Graduate Student Representative
    Iowa State University CS Committee (2022-2023)
  • 📚 Education Committee Member
    University of Vermont Graduate Senate (2019-2020)

🔬 Mentoring & Outreach

  • 🎓 Actively mentoring 9 students & researchers across Iowa State, BRAC, KIIT, IUB, and Amazon
  • ⚖️ Judge, State Science & Technology Fair of Iowa (2025)
  • 📖 Guest lecturer for AI/ML courses

🛠️ Technical Skills

💻 Programming & Development

Languages

Python    ⭐⭐⭐⭐⭐
Java      ⭐⭐⭐⭐
SQL       ⭐⭐⭐⭐
C++       ⭐⭐⭐⭐
R         ⭐⭐⭐
Go        ⭐⭐⭐

Web Development

FastAPI    ⭐⭐⭐⭐⭐
Flask      ⭐⭐⭐⭐
Streamlit  ⭐⭐⭐⭐
React      ⭐⭐⭐

Databases

PostgreSQL     ⭐⭐⭐⭐⭐
MySQL          ⭐⭐⭐⭐
MongoDB        ⭐⭐⭐⭐
Elasticsearch  ⭐⭐⭐⭐
SQLite         ⭐⭐⭐⭐
Athena         ⭐⭐⭐

DevOps & Cloud

AWS         ⭐⭐⭐⭐⭐
Docker      ⭐⭐⭐⭐
Git/GitHub  ⭐⭐⭐⭐⭐
Kubernetes  ⭐⭐⭐
Jenkins     ⭐⭐⭐
Terraform   ⭐⭐⭐

🧠 Machine Learning & AI

🔥 Deep Learning

  • PyTorch ⭐⭐⭐⭐⭐
  • TensorFlow/Keras ⭐⭐⭐⭐
  • Scikit-learn ⭐⭐⭐⭐⭐
  • ONNX Runtime ⭐⭐⭐⭐

🎬 Computer Vision

  • OpenCV ⭐⭐⭐⭐⭐
  • YOLO ⭐⭐⭐⭐
  • Detectron2 ⭐⭐⭐⭐
  • MMDetection ⭐⭐⭐

🤖 LLM & GenAI

  • LangChain ⭐⭐⭐⭐⭐
  • Ollama ⭐⭐⭐⭐
  • Azure Databricks ⭐⭐⭐⭐
  • TorchServe ⭐⭐⭐
  • Hugging Face ⭐⭐⭐⭐⭐

📊 Vector Databases

  • Chroma ⭐⭐⭐⭐
  • Faiss ⭐⭐⭐⭐
  • Pinecone ⭐⭐⭐

🎮 RL & Simulation

  • Ray/RLlib ⭐⭐⭐⭐⭐
  • Stable Baselines3 ⭐⭐⭐⭐
  • CARLA ⭐⭐⭐⭐
  • SUMO ⭐⭐⭐⭐
  • Isaac Gym ⭐⭐⭐
  • OpenAI Gym ⭐⭐⭐⭐

⚙️ Data Engineering & Analytics

🔄 Big Data Processing

Apache Spark      ⭐⭐⭐⭐⭐
PySpark          ⭐⭐⭐⭐⭐
Hadoop           ⭐⭐⭐⭐
Apache Kafka     ⭐⭐⭐
ETL Pipelines    ⭐⭐⭐⭐⭐
AWS Data Pipeline ⭐⭐⭐⭐

📊 Monitoring & Analytics

Splunk    ⭐⭐⭐
Datadog   ⭐⭐⭐

🗺️ GIS & Geospatial

QGIS                ⭐⭐⭐⭐⭐
ArcGIS              ⭐⭐⭐⭐
GDAL                ⭐⭐⭐⭐
PostGIS             ⭐⭐⭐⭐
Remote Sensing      ⭐⭐⭐⭐
Terrain Modeling    ⭐⭐⭐⭐
Spatial Analytics   ⭐⭐⭐⭐⭐

📈 Visualization

Tableau     ⭐⭐⭐⭐
Matplotlib  ⭐⭐⭐⭐⭐
Plotly      ⭐⭐⭐⭐
Seaborn     ⭐⭐⭐⭐
Kepler.gl   ⭐⭐⭐

🔬 Specialized Technologies

⚛️ Quantum Computing torchquantum
Qiskit
PennyLane
StrawberryFields

Novel QNN architectures for security applications

📊 Statistical Analysis R Studio
MATLAB
SciPy
StatsModels

Advanced statistical modeling and hypothesis testing

🎓 Research Tools LaTeX
Jupyter
Weights & Biases
MLflow

Experiment tracking and scientific writing

🏆 Domain Expertise

🔍 Crash Analysis          🎯 Prompt Engineering       ⏱️ Time Series Analysis
🤖 Autonomous Systems      📈 Predictive Modeling       🌐 Knowledge Graphs
🎥 Video Understanding     🧮 Optimization Algorithms   🔐 Cybersecurity

🎓 Education

🎓 Ph.D.

Computer Science
Iowa State University
Graduated October 2025

Dissertation Focus:
AI for Transportation Safety
& Video Understanding

🎓 M.S.

Artificial Intelligence
Iowa State University
Graduated January 2024

Specialization:
Machine Learning
& Computer Vision

🎓 B.Sc.

Computer Science & Engineering
BRAC University
Graduated August 2018

Focus Areas:
Software Engineering
& Algorithms


📊 GitHub Statistics



GitHub Streak



trophy


📫 Let's Connect!

💬 I'm always interested in collaborating on exciting AI and ML projects!


📧 Personal
ibnfarabishihab@gmail.com

🎓 Academic
ishihab@iastate.edu

📞 Phone
+1-347-571-4757

📍 Location
Sunnyvale, CA, USA


LinkedIn Google Scholar GitHub Email


🌟 Open to opportunities in:

AI Research • Machine Learning Engineering • Data Science • Academic Positions


Profile Views




⭐ Thank you for visiting my profile! Happy coding! 🚀

"I have to win. If not eventually, I will. There is no alternative."

Popular repositories Loading

  1. farabi_gone_autonomous farabi_gone_autonomous Public

    Python 3

  2. Online_Courses_Projects Online_Courses_Projects Public

    All projects from online courses I have taken with Coursera and Udacity

    Jupyter Notebook 2

  3. Scene-Graph-Benchmark.pytorch Scene-Graph-Benchmark.pytorch Public

    Forked from gladcolor/Scene-Graph-Benchmark.pytorch

    A new codebase for popular Scene Graph Generation methods (2020). Visualization & Scene Graph Extraction on custom images/datasets are provided. It's also a PyTorch implementation of paper “Unbiase…

    Jupyter Notebook 1

  4. algo_problem_solve algo_problem_solve Public

    This is the repository for all the problems I solved from Algo expert website

    Python 1

  5. datasciencecoursera datasciencecoursera Public

  6. datasharing datasharing Public

    Forked from jtleek/datasharing

    The Leek group guide to data sharing