Deepank Dixit
Verified Expert in Engineering
Machine Learning Developer
Bengaluru, Karnataka, India
Toptal member since September 24, 2024
Deepank has strong R&D and consulting experience in machine learning (ML), deep learning, and generative AI. At Blue Yonder, he designs and develops ML pipelines to enhance supply chain operations and optimize workflows through deep learning-based forecasting. He has expertise in generative AI, natural language processing (NLP), transformers, large language models (LLMs), SQL, Kubeflow, Snowflake, Azure, and MLOps, helping clients achieve scalable, efficient solutions.
Portfolio
Experience
- Machine Learning - 4 years
- Deep Learning - 4 years
- Keras - 4 years
- Retrieval-augmented Generation (RAG) - 4 years
- TensorFlow - 4 years
- Scikit-learn - 4 years
- Large Language Models (LLMs) - 3 years
- LangChain - 2 years
Availability
Preferred Environment
MacOS, Visual Studio, LangChain, Hugging Face, Scikit-learn, TensorFlow, Keras, Python 3
The most amazing...
...things I created were a deep learning-based pick forecasting system for warehouse operations and a VAE-based anomaly detection system in Cisco security Syslog.
Work Experience
Senior Data Scientist
Blue Yonder
- Built large-scale warehouse ML workflows using TFX, Kubeflow, Spark, and Snowflake to optimize logistics and supply chain operations.
- Developed an Agentic RAG-based multi-agent framework system to automate flowchart generation for warehouse operations, leveraging Llama 3.1 for long-context retrieval and dynamic action-pattern mapping.
- Used deep learning SOTA models for time series forecasting involving order and pick count.
Senior AI Engineer (Security)
Cisco
- Developed and led a semi-supervised generative modeling project to analyze and detect anomalies in Cisco ISE syslog data using VAE. Reduced the time required in log analysis by isolating the unusually large VAE losses.
- Built RAG and RAPTOR-based conversational AI assistants to allow for intuitive querying across various formats - from websites to documents to SQL databases- to offer users intelligent interaction with a vast array of data-driven tasks.
- Secured patented innovation through defensive publication for innovative use of variational autoencoders (VAEs) in Syslog anomaly detection with dynamic latent space adaptation.
Experience
Fully Local RAG with Ollama
https://github.com/DeepankDixit/Fully-Local-RAG-with-OllamaEducation
Master's Degree in Artificial Intelligence
Indian Institute of Science (IISc) - Bangalore, India
Bachelor's Degree in Computer Science
University of Petroleum and Energy Studies - Dehradun, India
Certifications
Advanced Retrieval for AI with Chroma
DeepLearning.AI
Multi AI Agent Systems with crewAI
DeepLearning.AI
Building and Evaluating Advanced RAG
DeepLearning.AI
Finetuning Large Language Models
DeepLearning.AI
LangChain Chat with Your Data
DeepLearning.AI
Generative AI with Large Language Models
Coursera
Skills
Libraries/APIs
OpenAI API, Scikit-learn, TensorFlow, Keras
Tools
Visual Studio, Git
Languages
Python, Python 3, SQL, Snowflake
Platforms
MacOS, Ollama, Kubeflow, Azure, Kubernetes, Docker, Vae
Frameworks
LangGraph
Paradigms
Anomaly Detection
Storage
Databases
Other
Artificial Intelligence (AI), Data Science, Machine Learning, Deep Learning, AI Consulting, Consulting, LangChain, Hugging Face, Programming, Computer Networking, Cybersecurity Automation, Retrieval-augmented Generation (RAG), Information Retrieval, Large Language Models (LLMs), Natural Language Processing (NLP), Data Analytics, Operating Systems, Data Structures, Algorithms, Security, ChromaDB, Multi-agent Systems, Fine-tuning, Probabilistic ML, Deep Representational Learning, Generative Modeling, Networking, Machine Learning Operations (MLOps), Transformers, GAN, Graph Neural Networks
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring