Hello, I am Priyanka Moorthy.

AI Engineer · Embodied AI & Multi-Agent Systems · 6 YOE

I build systems that learn from interaction — from cooperative LLM agents to robot manipulation policies.

Priyanka Moorthy
92%
routing accuracy
82%
faster doc processing
60%
claim throughput increase
22%
VLA loss reduction

About Me

I'm an AI Engineer with 6+ years of experience across the full ML stack — from NLP pipelines and distributed inference systems to multi-agent LLM architectures and Vision-Language-Action models for robot learning.

My work sits at the intersection of research and deployment: I've built cooperative agent systems that handle live healthcare operations at 92% routing accuracy and <2s latency, designed continuous evaluation frameworks with human feedback on Vertex AI, and trained a 77M parameter VLA model for multi-task robot manipulation using behavior cloning and cross-attention fusion. Earlier in my career, I built production NLP pipelines (OCR + CNN) that cut document processing time by 82% and designed asynchronous distributed inference systems with RabbitMQ and Celery to manage model concurrency at scale.

Across these roles, the through-line has been the same: take a well-defined problem, pick the right architecture, instrument it properly, and iterate until the system is reliable enough to depend on.

I recently completed a 77M parameter VLA model combining EfficientNet-B0 (vision) + DistilBERT (language) + cross-attention fusion to predict robot manipulation actions from camera images and natural language instructions. Using behavior cloning on scripted expert policies for pick/push/place tasks, I achieved 22% loss reduction through multi-task learning on Apple Silicon GPU with MPS acceleration. The full implementation is on GitHub.

RL & Robot Learning
Behavior Cloning VLA Models Multi-Task Learning MuJoCo PyTorch
Multi-Agent Systems
Function Calling Agent Routing Grounded Retrieval Agent Evaluation Gemini
Production ML
Vertex AI RLHF Pairwise Preference Model Monitoring Cloud Run
Deep Learning
Transformers Cross-Attention CNNs GANs LLM Fine-tuning
MLOps
Docker Kubernetes GCP RabbitMQ Celery
Healthcare AI
Claims Automation LLM-as-Judge Human Feedback de-id PHI/PII

I'm actively looking for teams working on real-world policy learning, sim-to-real transfer, or deploying RL agents in production environments — particularly in robotics, autonomous systems, or LLM-driven decision-making.

I'm most excited by roles where the research-to-deployment gap is the core problem: building reliable systems around learned behavior, designing evaluation frameworks for policies, and making models that generalize beyond their training distribution.

If your team is working on embodied AI, multi-agent coordination, or the infrastructure side of RL at scale — I'd love to connect.

priyankamoorthycit@gmail.com

Experience


AI Engineer @ Espercare LLC — Multi-agent LLM systems, Vertex AI pipelines, production ML evaluation
Jul 2023 – Present · San Jose, CA
Software Engineer @ RedBlackTree Tech — Production NLP pipelines, distributed inference, OCR + CNN classification
Jan 2019 – May 2020 · India

Projects


Reinforcement Learning
VLA Robot
Multi-Task Robot Learning
PyTorch MuJoCo EfficientNet-B0 DistilBERT Python

Built a 77M parameter Vision-Language-Action model combining EfficientNet-B0 vision encoder, DistilBERT language encoder, and cross-attention fusion to predict robot manipulation actions from camera images and natural language instructions. Achieved 22% loss reduction via multi-task behavior cloning across pick/push/place tasks.

Read More
DQN Mountain Car
Deep-Q Learning on Continuous Mountain Car
PyTorch OpenAI Gym Python

Implemented a sequential Deep-Q Learning agent on OpenAI's continuous mountain car environment, converging in 1643 episodes with 12 hidden dimensions.

Read More
Computer Vision
Pix2Pix Butterfly
Pix2Pix on Butterfly Data
TensorFlow Cycle-GAN OpenCV Python

GAN network for image translation from line art to colored butterfly images using a Pix2Pix architecture (U-Net Generator + PatchGAN Discriminator). Achieved discriminator and generator loss of 0.91 and 1.23 respectively.

Read More
Mask-RCNN Nuclei
Mask-RCNN on Nuclei Data
TensorFlow Mask-RCNN Python

Image segmentation model using Mask-RCNN for object detection and segmentation masks on nuclei microscopy images. Achieved a mean average precision of 0.73.

Read More
Sign Language Recognition
Continuous Sign Language Recognition
PyTorch OpenCV Python

Implemented a modified SAM-SLR framework for real-time continuous sign language recognition, integrating multi-cue visual streams for improved temporal modeling. Master's thesis project at SJSU.

Natural Language Processing
Abstractive Summarization
Abstractive Summarization
Pegasus Transformers BeautifulSoup Python

Scraped IMDB movie plot data, applied preprocessing (stop words, PorterStemmer), and fine-tuned a Pegasus model using a Self-Supervised Objective for abstractive summarization.

Read More
Sentiment Analysis
Sentiment Analysis on Amazon Reviews
DistilBERT LSTM CNN Python

Comparative analysis of sentiment classification on Amazon reviews, implementing and benchmarking FNN, CNN, LSTM, and DistilBERT models with word embeddings and data preprocessing.

Read More
Machine Learning
Lo-Fi Music Generation
Lo-Fi Music Generation
LSTM TensorFlow Python

Trained an LSTM model on Lo-Fi style music data to generate stylistically similar music sequences from learned temporal patterns.

Read More