Pavan Vemuri

CORE

Director of Product Engineering at SDVerse

Joined Jun 2024

About

Pavan Vemuri is a leading voice in the field of automotive AI strategy, software-defined vehicles, and MLOps platform development. With over 15 years of industry experience, he currently serves as Director of Product Engineering at SDVerse, where he shapes the company's AI strategy, builds cutting-edge AI capabilities for the automotive software marketplace, and leads initiatives in generative AI implementation, product adoption, and digital transformation. He has delivered expertise through his work at major automotive conferences and industry events, and continues to share insights on automotive AI implementation, MLOps for vehicles, and digital transformation strategies. Pavan holds an Executive MBA from Oakland University and an MS in Computer Science from Northern Illinois University. His unique blend of technical expertise and business acumen positions him as a trusted advisor to executives seeking to harness the power of AI innovation in the automotive sector.

Stats

Reputation:	1978
Pageviews:	129.6K
Articles:	25
Comments:	2

Articles
Comments

Articles

Essential Techniques for Production Vector Search Systems, Part 5: Reranking

Proven techniques for production vector search, including when to use each one, how to combine them effectively, and trade-offs to understand before deployment.

March 27, 2026

· 2,144 Views

Essential Techniques for Production Vector Search Systems, Part 4: Multi-Vector Search

Proven techniques for production vector search, including when to use each one, how to combine them effectively, and trade-offs to understand before deployment.

March 12, 2026

· 2,482 Views · 1 Like

Essential Techniques for Production Vector Search Systems, Part 3: Filterable HNSW

Proven techniques for production vector search, including when to use each one, how to combine them effectively, and trade-offs to understand before deployment.

January 30, 2026

· 1,508 Views · 2 Likes

Essential Techniques for Production Vector Search Systems Part 2 - Binary Quantization

Proven techniques for production vector search including when to use each one, how to combine them effectively, trade offs to understand before deployment.

January 9, 2026

· 1,910 Views · 3 Likes

Essential Techniques for Production Vector Search Systems Part 1 - Hybrid Search

Proven techniques for production vector search including when to use each one, how to combine them effectively, and trade offs to understand before deployment.

January 8, 2026

· 1,937 Views · 2 Likes

Why GPT-OSS:20B Feels Painfully Slow (And How Quantization Can Save Your Sanity)

GPT-OSS:20B's MXFP4 quantization requires H100+ GPUs. Use GGUF quantized versions from Hugging Face for dramatically faster consumer hardware performance.

October 6, 2025

· 5,275 Views · 4 Likes

Prompt Engineering Wasn't Enough; Context Engineering Is What Came Next

Expanded context windows enable a shift from prompt engineering to context engineering, i.e., optimizing what AI accesses rather than how we ask.

August 20, 2025

· 2,634 Views · 8 Likes

Talk to Your BigQuery Data Using Claude Desktop

Build an MCP server that lets Claude's desktop connect to your BigQuery data and converse with your data in natural language.

August 20, 2025

· 5,166 Views · 9 Likes

Agentic Workflows for Unlocking User Engagement Insights

Learn how to generate actionable insights by having agentic workflows deployed on your user engagement data with the help of CrewAI.

February 19, 2025

· 4,055 Views · 10 Likes

Knowledge Graphs and RAG: A Guide to AI Knowledge Retrieval

Enter knowledge graphs, the secret weapon for superior RAG applications. This guide has everything you need to begin leveraging RAG for intelligent AI knowledge retrieval.

Updated November 21, 2024

· 9,405 Views · 13 Likes

Decoding LLM Parameters, Part 2: Top-P (Nucleus Sampling)

The second article in this multi-part series explores how parameter tuning on Top-P impacts creativity, precision, and diversity in LLM content generation.

October 9, 2024

· 5,635 Views · 10 Likes

Decoding LLM Parameters, Part 1: Temperature

The first article in this multi-part series explores how parameter tuning impacts creativity, precision, and diversity in LLM content generation.

September 20, 2024

· 6,686 Views · 9 Likes

Automatic 1111: Custom Sketch-to-Image API

Develop a custom Sketch-to-Image API for converting hand-drawn/digital sketches into photorealistic images using stable diffusion models powered by ControlNet.

September 10, 2024

· 5,342 Views · 6 Likes

Order in Chaos: Python Configuration Management for Enterprise Applications

Explore a systematic approach to dealing with configurations in Python including tools, best practices, and real-world examples.

September 4, 2024

· 6,816 Views · 7 Likes

Better Search Results Through Intelligent Chunking and Metadata Integration

Chunking is an important, but underrated, aspect of LLM-based content retrieval, and effective management of It will yield world-class results.

August 26, 2024

· 6,104 Views · 8 Likes

Automatic 1111: Sketch-to-Image Workflow

With Stable Diffusion Web UI Automatic 1111, create custom workflows for converting Sketches into photo-realistic images using diffusion and ControlNet models.

August 21, 2024

· 7,195 Views · 7 Likes

Cloud Build Unleashed: Expert Techniques for CI/CD Optimization

Discover strategies to enhance speed and efficiency and ensure that your deployments are secure, compliant, and resilient.

August 19, 2024

· 9,572 Views · 9 Likes

Chat With Your Code: Conversational AI That Understands Your Codebase

In the ever-evolving software development landscape, interacting with your codebase conversationally can be a game-changer.

August 14, 2024

· 4,972 Views · 8 Likes

Automatic 1111: Adding Custom APIs

Following up on the last article that covered the fundamentals of Automatic 1111, let's take a look at the Options API and adding custom APIs.

July 31, 2024

· 4,599 Views · 7 Likes

Precision Python: Leveraging mypy and Pylint for Type Safety

Python's dynamic typing is flexible but error-prone. Static type checking with mypy and Pylint enhances code reliability and maintainability.

July 18, 2024

· 3,908 Views · 7 Likes

The Future of AI Art: Exploring the Capabilities of Stable Diffusion Web UI (Automatic 1111)

Automatic 1111's web UI for Stable Diffusion simplifies AI Image Generation: text-to-image, sketch-to-image, model training, and API integration detailed.

July 18, 2024

· 5,405 Views · 7 Likes

Poetry Explained: Perils of the Unpinned Dependencies

Poetry conquers Python's dependency madness with version pinning and conflict resolution, guaranteeing smooth, stable, and hassle-free releases.

July 17, 2024

· 6,005 Views · 8 Likes

Linting Excellence: How Black, isort, and Ruff Elevate Python Code Quality

In this article, explore Black, isort, and Ruff to streamline Python code quality checks and ensure consistent coding standards.

July 4, 2024

· 11,439 Views · 6 Likes

MLOps in Software-Defined Vehicles: A Centralized Platform Approach

MLOps platforms are essential for managing ML models in software-defined vehicles, ensuring their effectiveness and long-term success.

June 18, 2024

· 2,971 Views · 6 Likes

Unlocking the Power of Search: Keywords, Similarity, and Semantics Explained

An overview of keyword, similarity, and semantic search techniques, providing insights into how each one works and guidance on when to use them effectively.

June 13, 2024

· 6,406 Views · 8 Likes

Comments

Knowledge Graphs and RAG: A Guide to AI Knowledge Retrieval

Aug 21, 2024 · Pavan Vemuri

Thank you for the response. We will cover what you have asked in a future article.

DZone's Article Submission Guidelines

Jun 10, 2024 · Mike Gates

I am drafting my article, and when I am in preview mode, I see the article is cut off, and the final section is not showing up. I know the minimum limit is 500 words, but is there a maximum limit to the length of the article?