Leading a fantastic team of machine learning engineers and scientists to help companies build large language model powered solutions.

Amazon AI (Lex)- Conversational AI / dialogues machines.Research - Published research externally in top NLP conferences such as EMNLP and ACL, as well as internally, related to text generation using LLMs, meta learning, multilingual natural language understanding etc. In addition to this, supervised several PhD candidates from top universities during their internships at Amazon.Product Launches - Directly led the launches of Amazon Lex in over 5 languages serving several hundred thousand queries from customers worldwide, as well as developed systems for automated language launches, data augmentation using retrieval and generative techniques.

United Health Group (UHG) is the one of the biggest enterprises in the US. I worked with UHG’s Human Capital Research team and pursued various machine learning based models to help them better understand their enormous human resources (+200,000 employees). Here are a few of the many projects that I worked on:-- Talent network graph modeling and clustering similar employees based on skill sets-- Extraction of popular topics and employee sentiments via text mining using annual company surveys.-- Predicting resignation events for the next two years.

I have served as the Computation and Natural Sciences Lead Tutor, as well as the Subject Area Coordinator for Stanford’s Summer Session Tutor Program (STP) 2015. My job was to lead workshops and organize tutor sessions for both undergrad and grad students.Instructor and Designer of “Stanford - Machine Learning in Plain English” Workshop SeriesThis 4 day workshop series was designed by me so as to teach Stanford students machine learning algorithms using minimal math equations. This venture turned out to be one of the most popular series in the Stanford Summer Session 2015.

My MS research dealt with modeling the phase behavior and flow characteristics of fluidwhen it flows through nano-pores, often termed as nano-fluidics. At such a small scale, manyfluid properties change. My job was to study the multiphase flow behaviour that occurs in the nano-pores of shale reservoirs using a compositional simulator. I worked on enhancing Stanford’s in-house simulator ‘AD-GPRS' unconventional oil reservoir modeling capabilities.MS Thesis is available at-https://pangea.stanford.edu/ERE/db/pereports/record_detail.php?filename=haider2015.pdfResearch Supervisor - Dr. Khalid Aziz

Predicting shale wells' production potential using machine learning.