Hi, I’m
Aamir Ahmad Ansari
Data Scientist
New Delhi
Summary
Versatile Data Scientist specializing in Deep Learning, Machine Learning, Reinforcement Learning, and MLOps. Skilled in statistical and algebraic analysis, with expertise in developing robust models and end-to-end machine learning pipelines. Highly organized and self-motivated, delivering innovative solutions to complex data challenges. Committed to advancing AI technologies and driving meaningful insights in the rapidly evolving field of data science.
Overview
1
year of professional experience
3
Certifications
Work History
Roadzen Technologies
Data Scientist
7 2022 - Current
Job overview
- Implemented various GANs (VQGAN, StarGAN etc.) and diffusion models for generating synthetic image dataset to counter the lack of availability of image data for specific use cases
- Developed a one-stop robust framework to train, validate, infer and deploy (using FastAPI) image classifiers (including all classifiers available on Pytorch)
- Developed algorithms by incorporating image segmentation to clean large amounts of raw data to generate high quality labeled data to built several multi-class and multi-label classifiers for entry level data filtering and improving their accuracy by at least 10%
- Implemented end-to-end robust machine learning pipelines using modern data centric pipeline software, adding features like continual learning, data & model versioning, retraining, and experiment tracking
- Built advanced and intelligent chatbots using large language models to provide seamless customer experience. The bot provided answers based on the customers insurance documents
- LLM Hosting Framework: simplifies LLM usage for superior results, given the input and task - the framework dynamically generates prompts, automatically selects RAG optimization if required by the selected model, handles LLM interactions, and refines the output. It scales seamlessly with Kubernetes, ensuring consistently high performance.
Univ.AI
Consultant (MLOPS) (Part Time)
02.2023 - 12.2023
Job overview
- Developed robust python packages that standardize the code for building production grade machine learning pipelines with its automated experiment tracking (not offered with any other DAG creation library)
- The framework provides functionality of building DAGs with minimal code and retains the DAG information using mlflow.
- Tasks like data versioning and model versioning are provisioned to be automatically handled.
- It provides isolation in experiments as well as simple execution of the pipeline.
- The DAGs are compatible to be defined inside Jupiter-Notebooks and can be run using various orchestrators. The default that we ship with the package is Dagster, using which you can orchestrate notebooks and scripts.
Education
University of SouthamptonSouthampton, United Kingdom
Master of Science from Artificial Intelligence
09.2025
University Overview
An upcoming student at UoS September'24 Intake.
USICT, New Delhi
Bachelors of Technology (Electronics & Communication)
08.2017 - 6 2021
University Overview
GPA: 7.79
Gurusharan Convent, New DelhiDelhi, India
Senior Secondary Education
03.2016 - 3 2017
University Overview
- Final Grade: 85.3%
Skills
Python Programming
Machine Learning Operations(MLOPS)
Machine Learning
Deep Learning
Probability & Statistics
Neural Networks
Natural Language Processing
Analytical Skills
Data Visualization
Computer Vision
Reinforcement Learning
SQL
Birthcity
New Delhi
Birthcountry
India
Personal Information
- Date of Birth: 06/11/1998
- Nationality: Indian
- Marital Status: Unmarried
Projects
2023-12 - 2024-01, Dear Diary, New Delhi, India
- An implementation of a chat bot using Large Language Models and Streamlit which can act as your own personal digital diary.
- Integrated Google DeepMind's Gemini and Microsoft's Phi-2 models.
- Added SqLite Database for chat retention.
- Custom post-processing for Phi-2 outputs.
- User-friendly interface built with Streamlit.
2022-08 - 2022-12, Emotion Detection and Analysis, New Delhi, India
- It is a web application which helps digital public speakers to analyze their audience's behavior. It detects facial emotions, performs sentiment analysis on the speaker's speech and generates a complete report.
- The report contains time series analysis as intuitive graphs that shows the speakers what went right or wrong. We use SOTA models like YOLO-V7, Sentence BERT and OpenAI Whisper.
- Presentation on YouTube, starts at 16:15 mins. Youtube:https://www.youtube.com/watch?v=G5-u3YE2TSQ&t=931s | Github: https://github.com/aamir09/FinalSubmissionAI5
Certification
Masters AI & ML, Univ.AI
Publications & Research Work
2023-10 - 2024-05, Shape Morphing Evolution in Higher-Dimensional Space, New Delhi, India, (Remote, Part Time)
- Volunteering in Professor Ivo Sbalzarini's research group to track and cluster trajectories of various shapes that evolve over time.
- Spearheaded the development of cutting-edge data modeling techniques, consistently identifying and implementing optimal solutions for the given time series dataset using the following techniques: Elliptical Fourier Transform, Physics Informed Neural Networks, Neural ODE's, LSTMs.
- Conducted in-depth data analysis, extracting valuable insights using latent space analysis of the AutoEncoders fit on the data .
- Delivered weekly presentations featuring compelling data visualizations, effectively communicating findings and recommendations to supervisors.
2023-05, Publication, Tree based Convolutional Neural Networks for Image Classification, ICIDSSD'22
Timeline
Consultant (MLOPS) (Part Time)
Univ.AI
02.2023 - 12.2023
Masters AI & ML, Univ.AI
12-2022
Apache Spark, Edureka
02-2019
Big Data with Hadoop, Oracle
07-2018
USICT, New Delhi
Bachelors of Technology (Electronics & Communication)
08.2017 - 6 2021
Gurusharan Convent, New Delhi
Senior Secondary Education
03.2016 - 3 2017
Data Scientist
Roadzen Technologies
7 2022 - Current
University of Southampton
Master of Science from Artificial Intelligence