Parth Chokhra

Machine Learning Engineer

Pune, Maharashtra, India5 yrs 9 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Architected advanced ML solutions for drug safety monitoring.
Developed BERT-based models enhancing compliance efficiency.
Engineered scalable anonymisation pipelines for regulatory compliance.

Stackforce AI infers this person is a Healthcare-focused Machine Learning Engineer with expertise in NLP and compliance solutions.

Contact

Skills

Core Skills

Natural Language Processing (nlp)Large Language Models (llm)Machine Learning

Other Skills

Android StudioApache KafkaApache SparkC (Programming Language)Cascading Style Sheets (CSS)CommunicationData AnalysisData ScienceDatabase Management System (DBMS)Deep LearningDjangoHTML5JavaLeadershipMicrosoft Azure

About

As a Senior Machine Learning Engineer at Blue Altair, I have developed and deployed advanced machine learning solutions, including end-to-end pipelines on Azure for global drug safety monitoring and adverse event detection. My work includes building a BERT-based QA model, significantly improving compliance and operational efficiency, and creating a Retrieval-Augmented Generation (RAG) pipeline for mapping deviations to historical incidents, streamlining investigation timelines. I hold a Bachelor of Technology in Computer Science from the Indian Institute of Information Technology, Design and Manufacturing, Jabalpur. My expertise spans recommender systems, large language models, and PyTorch. I am committed to leveraging cutting-edge technologies to solve complex problems and optimise decision-making processes in the field of machine learning.

Experience

Blue altair

2 roles

Senior Machine Learning Engineer

Promoted

Apr 2023 – Present · 2 yrs 11 mos · Pune, Maharashtra, India · On-site

Architected and deployed an end-to-end ML pipeline on Azure for global drug safety monitoring. Built a BERT-based QA model for extracting route, dosage, and administration details from patient reports, powering weekly automated HTML reports for Teva products—reduced detection and reporting time by ~50%, improving compliance and monitoring efficiency.
Manufacturing Deviations – RAG & Agentic Systems: Developed a Retrieval-Augmented Generation (RAG) pipeline using Azure OpenAI + Azure Search to map new deviation cases to historical incidents, cutting investigation time significantly. Integrated LangGraph-based agentic workflows with SOPs and MBR guidelines, deployed via a chat interface, enhancing decision support and resolution speed across multiple global pharma plants.
Hidden Insights Mining from Unstructured Data: Collaborated with stakeholders to apply LangExtract and LLMs on patient testimonial data. Designed a proprietary “Surprise Signal” framework to surface hidden correlations during drug release cycles, enabling early risk detection and data-driven regulatory actions.
De-identification & Compliance: Engineered a scalable HIPAA/PII anonymisation pipeline capable of processing 100K+ records/week, ensuring compliance across regional regulations and seamless integration with Teva’s safety data infrastructure.

Natural Language Processing (NLP)Large Language Models (LLM)PyTorchApache KafkaRecommender Systems

Data Science Engineer - 1

Apr 2022 – Apr 2023 · 1 yr · Pune, Maharashtra, India · On-site

Designed and deployed an end-to-end ML pipeline on Azure for automating drug adverse event detection, integrated with an interactive dashboard for real-time regulatory evaluation.
Optimized the pipeline for low-latency, multi-user access, improving compliance workflows and enabling faster, more accurate safety signal detection.

Microsoft AzureNatural Language Processing (NLP)Python (Programming Language)Machine LearningSQL

Trell

2 roles

Data Science Engineer - 1

Jul 2021 – Apr 2022 · 9 mos · Bengaluru, Karnataka, India

Developed real-time content-based recommendation for Trell App users end-to-end, involving vector creation, storage, retrieval using milvus, and serving using Go API.

Data Science Intern

Sep 2020 – Jun 2021 · 9 mos · Bengaluru, Karnataka, India

Developed and deployed text generation systems on AWS, leveraging BigQuery for large-scale data storage and performance tracking to improve user engagement metrics.

Questgen.ai

Data Science Intern(Open-Source)

Apr 2020 – Aug 2020 · 4 mos · India

Worked on the transformer models mainly on BERT and GPT-2.
Build a question generation model using GPT-2 which can produce any number of the BOOLEAN question given a context or paragraph.
After GPT-2 now I tried working out current SOTA T5 transformers for all types of question generation. I am able to generate successfully generate general questions from the text.
Now aiming at generating multiple-choice options for a given text using transformers.