Kshitij Chawla

My joy is in building dependable software. I do my job so others can do theirs.

Technical Skills

1. Python, FastAPI, MongoDB, Docker, RESTful APIs, TDD
2. Flutter for Web, Microservices
3. Loggly, Datadog, Ansible, Linux, Circle CI, Azure Pipelines, Websockets, HTML, CSS, RBAC
4. AWS (EC1, EFS)

Backend & Frontend Developer & Manager
Duke University Health Systems

Durham, NC, USA 
Oct 2022 – Present

Flutter for Web, Python, FastAPI, Pydantic, AsyncIO, MongoDB, Docker, RESTful APIs, Microservices, Message Broker

Developed and managed a comprehensive event-driven web service from inception to execution, serving as both the sole creator and product owner. 

• Designed & implemented RESTful APIs using FastAPI & MongoDB, tailoring distinct API endpoints for each ML model.
• Established a resilient architecture based on a message broker, enabling quantum integration of microservices & swift deployment of new ML models & preprocessing protocols.
• Developed & deployed an intuitive GUI web application utilizing Flutter for Web, ensuring a user-friendly interface for interacting with the service.
• Optimized system performance, responsiveness, and resource utilization through asynchronous calls, multiprocessing, and CUDA.
   
• Comprehensively evaluated the tech stack and architectural choices; selected the most suitable solutions for the project's needs.
• Maintained ongoing & productive dialogues with stakeholders to discern product use cases & requirements, translating these insights into OKRs &  KPIs.
• Actively engaged with stakeholders through regular product demonstrations, fostering a collaborative environment and gathering valuable feedback to continually align the product with stakeholders' evolving requirements.

Backend Developer
Quant AI Labs

Paris, France
Feb 2022 – Sept 2022

Python, FastAPI, AsyncIO, MongoDB, Docker, AWS (EC1, EFS)

• Developed an in-house CRUD server utilizing FastAPI and MongoDB for efficient business data storage.
• Implemented Role-Based Access Control (RBAC) mechanisms to manage data access and permissions effectively.
• Successfully deployed the initial prototype on AWS EC2 and EFS infrastructure through manual procedures.
• Mentored junior developers, focusing on improving coding practices and enhancing documentation standards.
• Conducted a developer interview as part of the hiring process.

Backend Developer & Inner-Source Consultant
Quant AI Labs

Paris, France 
Feb 2022 – Sept 2022

Python3, Numpy, Pandas, Pytest

• Advised and supported the client in their initiative to transition to an inner-sourced approach for their analytics platform.
• Pioneered the establishment of contribution standards, governance models, & a structured decision-making hierarchy within the inner-source framework.
• Actively contributed by introducing new features and addressing issues in the in-house analytics platform.
• Enhanced and automated testing procedures through the implementation of Pytest, incorporating effective mocking and patching techniques.
• Elevated developer productivity by seamlessly integrating tools like pre-commit hooks into the development workflow.

Backend Developer
Trax Retail

Paris, France
Feb 2021 – Jan 2022

Python, FastAPI, Pydantic, MongoDB, AsyncIO, Redis, OpenCV, Docker, GCP

• Developed RESTful API service to efficiently assign data annotation tasks to our data quality team, improving productivity and accuracy.
• Implemented new image classification microservice utilizing OpenCV, enhancing the platform's capabilities for handling visual data.
• Implemented async Redis task queues using arq-redis for efficient backend processing, optimizing system performance and scalability.
• Wrote custom code for MLOps to effectively manage, version, and automatically benchmark ML models, streamlining model development.
   
• Successfully rescued a project from failure by addressing miscommunication issues among stakeholders, ensuring timely delivery and client satisfaction.
• Demonstrated strong project management skills by overseeing product development, gathering, and clarifying requirements, and establishing clear objectives and key results (OKRs).
• Holistically managed stakeholders by actively liaising across teams, facilitating discussions, and providing detailed credit attribution to foster teamwork and inclusiveness.

IoT Developer & On-Call Production Support
Trax Retail
Paris, France
April 2020 – Jan 2021

Python, Loggly, Datadog, Ansible, SaltStack, OpenCV, CUDA, concurrency, Linux, Docker, Websockets, ONNX

• Implemented a feature to save device state upon task interruption; then restore and retry.
• Built a web sockets based feature to report device parameters at high frequency.
• Improved device logging using Loggly and expanded metrics monitoring and alerting using Datadog.
• Automated the setup and deployment of our Docker app to our IoT devices using Ansible.
• Improved image and EXIF metadata capture and processing, and custom compiled libraries for CUDA.
   
• Proposed, initiated, and conducted 2 weeks long in-store observations to gather requirements, identify bugs, and address feature gaps.
• Provided after hours on-call support for our fledgling and not-yet-robust devices.
• Assessed solutions offered by a robotics B2B company and accordingly outlined potential future roadmaps.
• Assessed devices from an OEM with hands-on benchmarking, wrote code to make benchmarks reproducible and semi-automated; did some simple statistical analyses of the data.
• Embraced coding as a collaborative effort, actively onboarding and assisting team members while seeking support when needed.

Python Developer
INRIA Saclay
Palaiseau, France
July 2018 – Mar 2020

Python, Pytest, HTML, CSS, Circle CI, Azure Pipelines

Nilearn and Nistats are open-source neuroscience research libraries.    

• Integrated Nistats library into Nilearn, simplifying development and maintenance and improving user experience.
• Created functionality to easily replace function parameter names, handling replacement and deprecation, using decorators.
• Implemented a report generation and visualization feature written with HTML & CSS.
• Significantly improved and upgraded the continuous testing infrastructure to use Azure Pipelines, CircleCI.
• Migrated testing infra from Python2 to 3 and from Nose to Pytest.
• Improved API consistency, code maintainability, and documentation.
• Release management of 7 versions across the libraries 
   
• Ensured a good contribution and developer experience for our open-source contributors.
• Organized coding sprint in April 2019 in Paris and assisted in several Nilearn workshops.
• Gave an accepted suggestion during the neuroimaging data standards (BIDS) workshop (Aug ‘18) at Stanford University.
• Gave a talk about Nurturing an Open-Source Project at INCF Marseille, May 2019.

Masters in Biomedical Engineering