Skip to content
View NimaVahdat's full-sized avatar
28 followers · 16 following

Achievements

Achievement: YOLOAchievement: Pull Sharkx2Achievement: Pair Extraordinaire

Achievements

Achievement: YOLOAchievement: Pull Sharkx2Achievement: Pair Extraordinaire

Block or report NimaVahdat

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.

Content in all repositories owned by your account will be closed.
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
NimaVahdat/README.md

👋 Hey, I'm Nima Vahdat

I'm an AI Engineer and Researcher working at the intersection of machine learning, agentic AI, and intelligent systems.

My research focuses on building AI agents that learn from experience, retain decision context, and improve their behavior over time. I'm particularly interested in persistent behavioral learning, agent memory, and the engineering of reliable, adaptive AI systems.

My broader interests include:

  • 🤖 Agentic AI — autonomous agents, memory, and behavioral learning
  • 🧠 Machine Learning — deep learning, representation learning, and neural networks
  • 👁️ Computer Vision — graph neural networks and visual understanding
  • 🔬 AI Research — bridging research ideas with practical AI systems

I'm always open to connecting with fellow researchers and engineers, discussing ideas, and collaborating on interesting projects.

📧 Email: nimavah@yorku.ca

Pinned Loading

  1. decision-context-memory decision-context-memory Public

    Behavioral memory for LLM agents: compile corrections and tool failures into scoped RULE directives injected at the decision point. Five self-contained stages, 55% → 95% on DC-Bench.

    Python

  2. decision-context-benchmark decision-context-benchmark Public

    DC-Bench: a benchmark for behavioral memory — does an agent apply a past correction to an unseen task, not just recall it? 71 scenarios across 9 dimensions, built from real agent traces, with contr…

    Python

  3. contour-gnn contour-gnn Public

    Graph Neural Network framework for contour grouping, built on my thesis work. Provides tools to preprocess datasets (COCO & synthetic), build graphs, and train/test the NIVGraphNet model. 🚧 Work in…

    Python

  4. FoodSeg_mask2former FoodSeg_mask2former Public

    Fine-tuning Mask2Former on the FoodSeg103 dataset for food segmentation, with deployment and interactive inference via Gradio.

    Python 10 1

  5. DeepSNN-Framework DeepSNN-Framework Public

    DeepSNN Framework is an advanced tool for building Spiking Neural Networks, developed for deep learning applications. Tailored for modern neural networks, it features a flexible architecture, conv…

    Python 6

  6. ETH_Price_Prediction ETH_Price_Prediction Public

    Predict Ethereum (ETH) prices using LSTM and GRU neural networks.

    Python 6 1