Praharsh Dave
ML Engineer | AI & Generative Models | GPU Optimization (5.7× Speedups, 40%+ Latency Reduction) | MLOps & Cloud Deployments | LLMs, Transformers, LangChain
- Role
- Machine Learning Engineer at ServiceNow
- Location
- North Dartmouth, MA, US
- LinkedIn followers
- 500 followers
About Praharsh Dave
Machine Learning Engineer with 4 years of experience designing, developing, and deploying…
Experience
Machine Learning Engineer
Jan 2024 — Present · Dartmouth, MA, US
Accelerated inference for AI agents by optimizing large language models with ONNX Runtime and NVIDIA Triton Inference Server, achieving 40% lower latency in real-time IT service desk responses using GPU-based parallel processing. • Developed and deployed GPU-accelerated MLOps pipelines using CUDA and TensorRT, integrated with LangChain and AWS SageMaker, reducing model inference time by 30% for high-throughput workflows. • Developed Java-based microservices integrated with Spring Boot for AI workflow orchestration, enabling seamless communication between inference APIs and backend systems, reducing API response time by 15%. • Implemented Java-based data validation and transformation modules within the MLOps pipeline, ensuring 99% data integrity before model training and deployment. • Designed intelligent AI agents with LangChain and Lang Graph, leveraging Hugging Face Transformers for NLP tasks, resulting in a 25% reduction in ticket resolution time through enhanced intent detection and response generation. • Engineered real-time data processing pipelines with Apache Spark and Pandas to support AI agents built with Lang Graph, integrating with Snowflake and PostgreSQL for seamless data access, powering Tableau dashboards for operational insights. • Fine-tuned deep learning models using PyTorch and LangChain for predictive maintenance and workflow automation, deploying via FastAPI and REST APIs to production systems, boosting process efficiency by 20%. • Designed and deployed multi-agent systems with AutoGen to orchestrate complex IT operations, leveraging TensorFlow and Scikit-learn for predictive analytics, improving incident prioritization accuracy by 18% in high-volume environments.
Education
University of Massachusetts Dartmouth
Master's degree
2022
ADITYA SILVER OAK INSTITUTE OF TECHNOLOGY
Bachelor's degree
2016 — 2020
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.