VS
02 / 06Who I am

One engineer, three overlapping problems.

I build AI systems, data pipelines, and the production software around them — then measure whether they actually work.

Profile01

I am a Master's student in Information Management at the University of Washington, where I hold a 4.0 GPA, and a text mining specialist at UW Libraries and the eScience Institute. My work sits across three things that are really one thing: the AI system, the data underneath it, and the software that has to keep both running.

Before Seattle I spent a few years across machine learning and full-stack engineering, building systems end to end — collecting data, processing it, training models, and putting them into production with the APIs, infrastructure, and CI around them. These days I am most drawn to agentic AI, and to the harder half of it: proving a system got better rather than asserting it did.

Education02
University of Washington, Seattle
M.S. in Information Management·Expected June 2027·GPA 4.0
Acropolis Institute of Technology and Research, India
B.Tech in Computer Science Engineering·June 2025
Vedant Somani speaking on stage
Speaking at a campus tech event
Toolkit03

Languages & Databases

  • Python
  • C++
  • C#
  • Java
  • JavaScript
  • TypeScript
  • HTML5
  • CSS3
  • MySQL
  • PostgreSQL
  • MongoDB
  • DuckDB
  • Pinecone

AI & Agents

  • Claude Code
  • Cursor
  • GitHub Copilot
  • OpenAI GPT-4o
  • Gemini
  • NVIDIA NIM
  • Ollama
  • LangChain
  • LangGraph
  • LangSmith
  • LlamaIndex
  • CrewAI
  • AutoGen
  • Haystack
  • MCP
  • Hugging Face
  • RAG Pipelines
  • Multi-Agent Systems
  • Tool Calling
  • BERT

Evaluation & Observability

  • LLM Evals
  • RAGAS
  • DeepEval
  • Phoenix
  • OpenTelemetry
  • MLflow

Machine Learning & Data Science

  • PyTorch
  • TensorFlow
  • Scikit-learn
  • XGBoost
  • NumPy
  • Pandas
  • SciPy
  • Matplotlib
  • ONNX
  • OpenCV
  • Causal Inference
  • Apache Spark
  • PySpark
  • Apache Kafka
  • Hadoop
  • Snowflake
  • Scala
  • MATLAB
  • Streamlit
  • Tableau

Software & Web

  • FastAPI
  • Django
  • React
  • React Native
  • Next.js
  • Node.js

Platforms & Tools

  • AWS
  • Azure
  • Docker
  • Kubernetes
  • Linux
  • Git
  • GitHub
  • GitHub Actions
  • Postman
  • Jira
  • Figma
Recognition04

Awards

Dempsey Startup Competition 2026

Investment round finalist, top 40 teams at UW Foster.

Read the story

Instructed at the AI in Practice Summer Institute

Led the sequence-models track — RNNs, LSTMs, and transformers — for researchers, postdocs, PhD students, and faculty at the UW eScience Institute.

Course site

iEngage Hackathon

Runner-up at the Salesforce and Lovable hackathon. Built Atlas, an agentic student success portal.

Try the project live

UW AIMS Product Competition

Runner-up, Association of Information Management Students at UW.

View the project

Publications

IEEE ACROSET 2025

PyFao-driven precision irrigation and fertilizer optimization in agriculture.

Read on ResearchGate

IEEE ICTBIG 2024

Elderly well-being through volunteer engagement.

Read on ResearchGate

Certifications

Fundamentals of Deep Learning

NVIDIA certification.

AWS Academy Graduate

AWS Academy Cloud Architecting.

Community & talks05

Beyond the commit log: hackathons I have judged, workshops and institutes I have taught, and the communities that shaped how I build.

Instructing the sequence-models and transformers sessions at the AI in Practice Summer Institute (photo 1 of 5)

Aug 2026

Instructing at the AI in Practice Summer Institute

Five days, 9 to 5, in a room of researchers, postdocs, PhD students, and faculty working through what it actually takes to build and evaluate AI models on their own problems. The sequence track was mine to lead: RNNs and LSTMs on time series, then transformers and the attention mechanism, built up from why a fixed context vector runs out of room, through query, key, and value, to the full scaled dot-product. Thanks to Valentina Staneva, Anthony Arendt, and the eScience team for putting it together.

TeachingTransformersUW eScience
LinkedIn
Vedant with his Dempsey Startup Competition team

May 2026

Investment round finalist at the Dempsey Startup Competition

Our team took ResolveIQ to the investment round, the top 40 teams, at UW Foster's Dempsey Startup Competition run by the Buerk Center for Entrepreneurship. It taught me a lot that goes well beyond building a product and pitching it, alongside Harsh Vardhan, Devika Bhutani, Sanchit Arora, and Sunayana H.

StartupPitchingUW Foster
LinkedIn
Team Forde 102 at the UW AIMS Product Competition

Mar 2026

Runner-up at the UW AIMS Product Competition with Marcus

We built a working prototype of Marcus, a trust-first campus marketplace with verified identities, TrustScores, and AI-driven fair pricing, in six hours. It was grounded in real user research done on campus rather than conference-room guesses. Marcus is still a prototype, but the thinking behind it is something I want to keep building on.

ProductUser ResearchUW AIMS
LinkedIn
The Atlas team at the iEngage hackathon holding their project poster (photo 1 of 2)

2026

Runner-up at the iEngage Salesforce and Lovable hackathon

Built Atlas, an all-in-one student success agent that connects every campus service through a single conversation, at the iEngage hackathon with Salesforce and Lovable. It took runner-up and grew into a full project with a Gemini-powered agent, Salesforce integration, and a RAG knowledge base.

HackathonAgentforceAtlas
Try Atlas live
Judging and mentoring at the WebSprint hackathon (photo 1 of 3)

Apr 2025

Judge and mentor at WebSprint, GDG on Campus IET-DAVV

Judged and mentored at WebSprint, a web-development hackathon with more than 60 teams. Reviewing that many builds in one day sharpened my sense of what separates a good demo from a real product, and it felt good to give back to the student community that got me started.

MentorshipHackathonCommunity
LinkedIn
Delivering the AICTE FDP workshop on AI-enabled structural health monitoring (photo 1 of 4)

Dec 2024

AICTE FDP workshop on AI-enabled structural health monitoring

Led a week-long hands-on workshop for an AICTE Faculty Development Program on AI-enabled structural health monitoring, covering anomaly-detection models, IoT sensor networks, and real-time data analytics for infrastructure safety, working with faculty and researchers.

AIIoTTeaching
LinkedIn

Apr 2024

SmartSculpt 2.0, a hands-on IoT workshop at AITR ACM

Taught at SmartSculpt 2.0, an IoT workshop run by the AITR ACM chapter where students learned the fundamentals and built their own working cars, alongside fellow speakers Tanay Patil, Somnath Tyagi, and Satwik Tiwari.

IoTWorkshopACM
LinkedIn
The Prayatna Hackathon organized by the AITR ACM chapter (photo 1 of 3)

Apr 2024

Organizing the Prayatna Hackathon as AITR ACM Chairperson

Led the AITR ACM Students Chapter in running Prayatna, a hackathon focused on technology for social challenges. It brought developers, designers, and builders together with more than 20 industry mentors and judges, plus chief guests from DigiSparsh and HotWax Systems.

LeadershipHackathonTechForGood
LinkedIn
Vedant at the GDG DevFest 2022 sponsor wall in Indore

2022

GDG DevFest 2022, Indore

Where a lot of it began. Attending Google Developer Groups' DevFest in Indore, taking in the talks and meeting the community that later shaped my own work running hackathons and workshops.

CommunityGDGDevFest