Skip to content
View dhvani-k's full-sized avatar

Block or report dhvani-k

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
dhvani-k/README.md

Dhvani Kothari

AI Solutions Architect | Forward Deployed Engineer | Applied AI Field Engineer

LinkedIn Email

I work with enterprise customers to get AI systems into production. Most projects start with a loosely defined business problem and a lot of existing infrastructure. My job is to figure out what should actually be built, build it, connect it to the customer's data and systems, and stay involved until their team is using it day to day.

I came into this from data engineering, so I spend more time than most on the data and integration side of AI projects. In my experience that's where things usually break.

Customer problem → Discovery → Architecture → Prototype → Data → AI → Integration → Production → Adoption

What I work on

  • Technical discovery with customers: understanding the problem, the data they have, their constraints, and what "working" means to them
  • Architecture and solution design for AI systems that need to fit into an existing enterprise environment
  • Prototypes and proofs of concept, and turning the ones that work into production deployments
  • RAG systems, LLM applications, and agent workflows, mostly on Claude and OpenAI models
  • Prompt engineering and evaluation. I build eval sets from real usage so prompt and model changes don't quietly break things
  • Data pipelines, migrations, and validation to get customer data into a usable state
  • Integrations with the APIs and systems customers already run
  • Troubleshooting across the stack when something in production isn't behaving

Stack

AI / GenAI LLMs, RAG, AI agents and agentic workflows, prompt engineering, LLM evaluation, MCP, Guardrails, Claude, OpenAI, Cursor
Data Python, SQL, ETL / ELT, data pipelines, data migration and transformation, data validation and quality, large-scale data processing, relational databases
Cloud AWS, Azure, cloud architecture, data lakes, APIs, system integration, enterprise architecture, on-prem to cloud migration
Solutions work Solution architecture, technical discovery, requirements gathering, proofs of concept, enterprise deployments, troubleshooting, customer adoption, working across engineering, product, and customer teams

Background

Before moving into AI solutions work I was a data engineer at Walmart Global Tech (ETL on GCP, PySpark pipelines, data migration) and at Capgemini (Spark, data lakes, Azure Data Factory automation). At BrainChip I was an ML Solutions Architect working on LLM fine-tuning, RAG pipelines, and voice solutions for customers. MS in Computer Science and Engineering from the University at Buffalo.

Projects

I'm in the process of putting some of my own work up here. Nothing worth pointing at yet.

Contact

Pinned Loading

  1. Car_Rental_Database_Design Car_Rental_Database_Design Public

    Car Rental Database Design using PostgreSQL

    PLpgSQL 7

  2. Firearm_Detection_In_CCTVs_Using_YOLOv5 Firearm_Detection_In_CCTVs_Using_YOLOv5 Public

    Detects firearms in real-time from CCTV videos, highlighting them within bounding boxes for immediate attention

    Jupyter Notebook 9 2

  3. YouTrend_Insights_Analyzing_YouTube_Video_Landscape YouTrend_Insights_Analyzing_YouTube_Video_Landscape Public

    An end-to-end solution for managing and analyzing YouTube video data from Kaggle, leveraging AWS services and visualized through Quicksight and Tableau

    Python 1

  4. CineETL_Movie_Insights_Data_Pipeline CineETL_Movie_Insights_Data_Pipeline Public

    A data pipeline that conducts ETL processes to AWS Redshift, utilizing Spark and coordinated by Apache Airflow.

    Python

  5. NLP_Sentence_Analysis_Feature_Extraction NLP_Sentence_Analysis_Feature_Extraction Public

    Code for sentence analysis and feature extraction using NLP

    Python

  6. NYC_311_Service_Insights NYC_311_Service_Insights Public

    NYC-311 Service Insights: A data-driven analysis of NYC's non-emergency service requests from 2010 to 2023

    Python