Shahbaz Khan Email & Phone Number

Munich, Bavaria, Germany

Tweaked diffusion models for anomaly detection

Munich, Bavaria, Germany

As a HiWi Tutor, assisted the chair of Data Analytics and Machine Learning in the correction of the following exam(s) for 3 consecutive semesters:1. Machine Learning for Graphs and Sequential Data (IN2323)2. Machine Learning (IN2064)

Munich, Bavaria, Germany

Worked part-time since the company's inception with the founding team to develop the first prototypes and the current platform that is being used by paid customers

Munich, Bavaria, Germany

Investigated shortcut learning in deep neural networks.Designed a novel method based on variational autoencoders to detect shortcuts in ML datasets.Published at the ICML 2023 Workshop on Spurious Correlations, Invariance, and Stability.

Munich, Bavaria, Germany

Built a real-time instance segmentation model for a fleet of delivery robots.Designed a novel architecture to train a model with incompletely labeled training data.

Munich, Bavaria, Germany

Developed sales prediction models for food retail in Germany.

Mumbai, Maharashtra, India

Developed ETL pipeline on AWS to fetch and process large-scale data (~600 million rows) from multiple channels. Built a marketing analytics dashboard using this data.

Mumbai, Maharashtra, India

• Developed deep learning based image captcha solvers for the data engineering pipeline that helped the company gather data from a wider set of data sources.
• Built OCR pipelines for extracting information from bank cheques, tax filings, KYC docs, etc.

Kanpur, Uttar Pradesh, India

• Developed a novel approach to identify genuine identity markers on product labels.
• Machine Learning on Image features: Extracted important features from images of physical anti-counterfeit tags and trained ML models on the same, deployed it on Android through Java Native Interface (JNI). The training.

Mumbai Area, India

• Credit Risk Evaluation: Developed a credit-scoring model for finance B2Micro lending platform using tree-based ensembles. Developed Gradient Boosted Tree models, used Grid Search with Cross-Validation to find the best.
• KYCification: POC for information extraction from KYC documents. Used Mask-RCNN to segment PAN cards from images of xerox copies, applied OCR to extract details from segmented images.
• Doctor’s Glance: Designed user interfaces (ReactJS), Developed REST APIs (Django), implemented global state management in Redux, designed image annotator app for a large chain of dermatology clinics and deployed as a.
• Justice League India: Performed data munging tasks for ongoing annotations of legal data, wrote SQL queries and worked with Pandas DataFrames

• Assisted students in the Python course: solved queries, helped in assignments, final project and evaluated coursework
• Beta tested the course and submitted feedback for improvement.

Represented the brand in the student's community.Was ranked 30 from among 11000+ student partners across the globe.Also won the 'ISP 9.0 Trainings Gold Award' for getting record enrolments in Internshala's training courses.

• Developed an Alexa skill that responds to queries related to IPL tournament (http://bit.ly/2WzLMrB).
• Coded the HTTP endpoint API of the skill in Python Flask, with PostgreSQL, deployed on Hasura.
• Configured & deployed the skill on AWS.

Actively promoted and generated large number of enrolments for the course IMAD (Introduction to Mobile Application & Development), offered by IIT Madras jointly with Hasura. Listed as top 10 performers from among 1400+ Student Ambassadors.

• Image API Project: Wrote Python scripts to crawl and extract images and their meta-data from fashion e-commerce websites. Used BeautifulSoup library for HTML parsing.
• Extracted and processed 6 GB+ of image & text data.
• Fetched data comprised of 30000+ images across 30+ categories.

Mumbai Area, India

• Worked along with Big Data and Data Science teams.
• Manually annotated datasets for image segmentation, objection detection and action analysis, which are used to build Deep Learning Algorithms.
• Semi-automated most of the tasks by writing scripts in Python.

Master Of Science - Ms, Computer Science

Bachelor Of Engineering (B.E.), Computer Engineering

Activities and Societies: 1. Led the technical team of CSI students chapter in college 2. 2nd Runner-up at the national-level Hackathon.

Higher Secondary School Cetificate (Hsc), Computer Science, Distinction