Haider Ali

Creating a harmful prompt detector using llm from scratch using JAX

Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, VLLM, NVIDIA NIM]

Balanced dataset using generated images from Stable Diffusion

This project aims to launch freebies posted on the day posted by Doctor Of Credit

An AI agent fetches email promotions and notifies you about them according to your preferences so that you don't miss on a deal

My portfolio page

Converting resume to word cloud

This tool tracks Gold and Silver ETFs and USD to INR using API and notifies

Multi-Agent AI Chatbot to assist with activate, deactivate card. Also, answers any card related questions.

Local RAG pipeline leveraging open source models

Here is a list of all the tech community slacks!

Predicting Medical Expenses from US data.

This was the final year project of my BSc Computer science degree.

A RAG app for answering questions using Langchain, Gradio, Ollama and Groq

🧙 Build, run, and manage data pipelines for integrating and transforming data.

Detecting Deep Fake Faces using Transformer and CNN model

Create a fire portfolio in 3 mins! Connect LinkedIn, GitHub, or resume & let AI magic happen.

Code files for data preparation and fine tuning models

An application tracking system leveraging ChatGPT

This project aims to classify brain signals for identify different body movements

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

Sleep state detection of child based on sleeping angles and movement

Classifying different kind of protein cells.

Question generation using state-of-the-art Natural Language Processing algorithms

Deep Q-learning driven stock trader bot

No description provided.

Using Image processing semgenting ALL images

matplotlib: plotting with Python

In this project I've used SIFT, SURF and HOG feature extractors to classify images into 1, 2, 3, and 4 classes 1 being lowest keratosis level and 4 being the highest level using KNN, SVC models.

Segmentation of lung sections from chest images