Skip to content
View N-G-Asker's full-sized avatar

Block or report N-G-Asker

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
N-G-Asker/README.md

Nicholas Asker
Software Engineer focused on AI and Computer Security
MS in Computer Science, Columbia University
Machine Learning track
[email protected]

Research Interest

My current research interest is in LLM-powered autonomous agents.

I am especially excited by the prospect of improving model performance and decision-making through

  • retrieval-augmented generation (RAG): providing relevant information as context using intelligent search techniques (using vector- or embedding-indexed databases), and

  • self-improvement and automatic feedback mechanisms such as self-refine.

Background

I like coding because it lets me find flow, be curious, and create. I'm interested in machine learning in particular because of its potential to solve otherwise intractable problems.

I recently graduated with a masters in computer science from Columbia University (May 2024). I chose the Machine Learning track as my specialization and am especially passionate about Large Language Models (LLMs) and deep learning.

My first exposure to coding came while I was a cybersecurity consultant (2017 - 2021), where I had the opportunity to pick up Python. I felt great joy learning to program for the first time and suddenly realized how much constructive power a programmer has – literally at their fingertips – with the capacity to create highly useful and beautiful things through code. So, I committed to going back to school to study computer science – to build a firm engineering foundation and pivot my career into something I loved.

The first half of my graduate program consisted of fundamental coursework in data structures, algorithms, systems programming, and mathematics. Through this portion of the program, I became a proficient programmer in C, Python, and Java and developed strong problem-solving skills. In the latter half, I shifted focus to topics in Artificial Intelligence $-$ including Natural Language Processing (NLP), Language Generation, Computer Vision, and Robot Learning $-$ and conducted research in these areas:

  • TasteRank: Personalized Image Search and Recommendation [Paper]

  • Whiteboard then Code: Refining Code Generation via Iterative Automatic Feedback from Peer LLMs [Paper]

Pinned Loading

  1. whiteboard-then-code whiteboard-then-code Public

    An inference-time framework for code-generation tasks, pairing content-planning ("whiteboarding") with simulated collaboration from peer LLMs ("team-refine").

    Python 1

  2. TasteRank TasteRank Public

    TasteRank: Personalized Image Search and Recommendation. This research project proposes an AI-based method for scoring photos on relevance to user interests. TasteRank leverages language and vision…

    Jupyter Notebook 1