Tejaswini Bharatha

Passionate about Machine Learning, Computer Vision, Generative AI, and NLP with a robust academic background and professional experience spanning various domains of artificial intelligence. Building intelligent systems at scale.

3+
Years Experience
ML/AI
Specialization
NYC
Location

Experience

ML Engineer

Peloton Interactive
Jan 2025 – Present
  • Built and productionized large-scale Whisper ASR + translation pipelines for multilingual subtitles (EN/DE/ES), cutting vendor costs by $10,000+/month with automated training, evaluation, and deployment via Airflow DAGs
  • Designed and deployed personalization models dynamically adapting ride resistance and cadence per user using historical performance and engagement data
  • Developed AI-powered class description generation systems leveraging LLMs and structured prompting for high-quality class metadata
  • Designed and deployed end-to-end MLOps workflows using ArgoCD and Datadog for production reliability

Founding Engineer

Ace AI
Nov 2024 – Jan 2025
  • Built foundational AI infrastructure and backend systems for early-stage AI platform
  • Architected scalable machine learning pipelines and data processing workflows
  • Developed core AI models and deployment systems for production environments

AI Backend Engineer

MealMatch AI
Oct 2024 – Jan 2025
  • Engineered AI-powered backend systems for personalized meal recommendations
  • Developed machine learning pipelines for nutritional data processing and user preference analysis
  • Built APIs and services for real-time recommendation delivery at scale

AI Engineer

Bear Brown & Company
Sept 2024 – Dec 2024
  • Developed state-of-the-art Vision Transformer (ViT) recommendation system capturing fine-grained visual features like product patterns, textures, and brand-specific details
  • Leveraged DeiT to reduce training time and computational cost while maintaining performance
  • Integrated ViT with GPT-4 and BERT for cross-modal recommendations, achieving 22% increase in CTR

ML Engineer (Computer Vision)

Tatum Robotics
Sept 2023 – May 2024
  • Developed ASL sign segmentation system using MediaPipe, collecting and computerizing dataset from scratch to enhance accuracy
  • Engineered real-time ASL to text translation system with CNN-GRU model, improving communication for 500+ deaf-blind users with 3x interaction efficiency gains
  • Led full-cycle deployment of AI model with T1 hand robot using socket programming for TCP message exchanges
  • Deployed system using FastAPI for real-time processing, enabling seamless robotic response to ASL

ML Engineer

Frigg Business Solutions
April 2021 – April 2022
  • Architected computer vision system detecting template presence in 10K+ product images monthly, reducing manual inspection by 70 hours
  • Conducted A/B testing to validate system effectiveness and optimize detection algorithms
  • Established OCR + NLP solution extracting nutritional data from food labels, automating catalog updates for 5K+ SKUs with 40% completion speed increase
  • Developed Kafka-based data pipelines processing 500GB+ datasets, reducing processing time by ~2 hours
  • Orchestrated AWS/Azure migration saving $20K annually through optimized resource utilization

Computer Vision Intern

Fraternity of Industrial & Automobile Engineers
Dec 2020 – Feb 2021
  • Designed driver vehicle integration systems using TensorFlow, minimizing offset crashes by 30%
  • Employed CNN + LSTM model achieving 30% increase in vehicle range and 20% decrease in charging time
  • Developed Random Forest predictive maintenance model reducing unexpected breakdowns by 25%
  • Received recognition for leading top-performing project on driver vehicle systems for electric two-wheelers

GitHub Projects

Loading your repositories...

Skills & Expertise

Core Competencies

Deep Learning NLP Computer Vision Generative AI Reinforcement Learning A/B Testing Data Science

Languages

Python Java C/C++ SQL (Postgres) JavaScript Go R

Frameworks & Tools

FastAPI PyTorch TensorFlow AWS SageMaker GCP Docker Airflow ArgoCD

Libraries & Tools

pandas OpenCV MediaPipe scikit-learn YOLOv5 PySpark Kafka

Testimonials

⭐⭐⭐⭐⭐
"Tejaswini worked at Tatum Robotics for a co-op and continued her research with us to develop fingerspelling gesture recognition software. She went above and beyond, impressing our team with her work ethic and competency. By the end of her time on our team, her individual and collaborative work had transformed our earlier models, showing promise and efficacy going forward. I highly recommend Tejaswini to computer vision teams!"
Tatum Robotics
Computer Vision & Robotics

Certifications & Awards

Education

MS in Robotics, Computer Science
Northeastern University, Boston • Jan 2022 – May 2024
Coursework: Reinforcement Learning, Algorithms, Machine Learning & Pattern Recognition, Data Visualization, Robot Mechanics & Control, Assistive Robotics, Robot Sensing & Navigation
BS in Mechatronics, Computer Science
Mahatma Gandhi Institute of Technology, India • May 2017 – May 2018
Focus: Robotics, Control Systems, and Computer Vision

Get in Touch