University of Michigan, Ann Arbor
M.S. in Computer Science and Engineering
Aug 2026 – May 2028 (expected) · Ann Arbor, MI
Fall 2026 coursework
- Advanced Scalable Systems
- Advanced Compilers (CSE 583)
- Natural Language Processing
Open to Summer 2027 internships · US · May–Aug 2027
Sri Satya Vinay Venneti
MS CSE @ University of MichiganEx-Software Engineer @ Wells Fargo
I build AI agents and the backend systems they run on. At Wells Fargo I built a multi-agent AI platform that cut campaign time-to-market from 6 months to under 2 hours. I'm looking for a Summer 2027 Software Engineering / AI Engineering internship.
About
About me
“The more I learn, the more I realize how much I don’t know.”
I like building systems end to end, from the data pipelines underneath to the AI agents on top. I’m now doing my MS in Computer Science and Engineering at the University of Michigan, graduating May 2028.
In two years at Wells Fargo I worked on both layers: first modernizing a terabyte-scale customer-data pipeline with Spark, then building a multi-agent AI platform with Google ADK, MCP and A2A that turns a marketer’s intent into a launch-ready campaign.
At Michigan I’m porting a declarative computer-use interface to macOS agents, and my team took 2nd place in the Agentic AI track at JacHacks 2026.
Before that: a B.Tech with Honors from IIIT Hyderabad (9.18/10), blockchain research at CSTAR Lab, and TA work for Introduction to Information Security.
When I'm not coding
- Binge-watching series
- Travelling
- Reading the news
- Swimming
- Singing
- Table tennis
By the numbers
6 mo → <2 hrs
Campaign time-to-market with the multi-agent AI platform
90%
Faster dataset provisioning (6 hours → 30 minutes)
4×
Microsoft Azure certified
2nd
Agentic AI track, JacHacks 2026
Fun facts
Favorite place
Kodaikanal
Typing speed
80+ wpm
Preferred search engine
Bing, yes Bing
Currently reading
The Berry Pickers by Amanda Peters
Favorite food
Gutti Vankaya (Brinjal Curry)
Fuel of choice
Just plain milk
Soundtrack
Movie scores
Experience
Where I've worked
Two years shipping AI and data systems at Wells Fargo, plus internships and research before that.
Wells Fargo
Hyderabad, India
Software Engineer · Marketing Technology (MarTech)
–
- Built a multi-agent AI marketing-orchestration platform on Google ADK, one of Wells Fargo’s top-3 AI initiatives, cutting campaign time-to-market from 6 months to under 2 hours. Earned a Team Spotlight Award.
- Designed specialized LLM agents for audience targeting, personalized content, offer curation, and automated compliance and brand checks: a marketer states campaign intent, the platform returns a launch-ready campaign.
- Integrated third-party CRM vendor agents and tools over the A2A protocol and Model Context Protocol (MCP).
- Built agent grounding and observability: a knowledge base, lifecycle-event monitoring, Arize Phoenix tracing of every reasoning step and tool call, and a golden evaluation dataset for regression testing.
- Shipped containerized backend services and REST APIs to Red Hat OpenShift through GitHub Actions and Harness CI/CD.
- Python
- Google ADK
- LLMs
- MCP
- A2A
- Arize Phoenix
- REST APIs
- Azure
- OpenShift
- GitHub Actions
- Harness
Program Associate
–
- Cut dataset provisioning time by 90% (6 hours → 30 minutes) with Python and Spark automation.
- Modernized a terabyte-scale legacy data pipeline with offer-code processing and real-time alerting, so eligibility checks ran on 100% of the latest qualified customer data.
- Built ETL workflows feeding a Unified Customer Profile in Adobe Experience Platform, the single source of truth for campaign eligibility.
- Apache Spark
- Spark SQL
- Python
- ETL
- SQL
- Adobe Experience Platform
Remedo Clinitech
Remote
Backend Developer Intern
–
- Wrote unit tests for the core TypeScript/NestJS backend that exposed 4 critical bugs, all fixed.
- Integrated SQLite into NestJS’s built-in test framework, cutting test runtime by 60%.
- TypeScript
- NestJS
- Node.js
- Jest
- TypeORM
- SQLite
CSTAR Lab, IIIT Hyderabad
Hyderabad, India
Research Associate
–
- Designed the blockchain architecture and Hyperledger Fabric smart contracts for a MeitY-funded (Government of India) privacy-preserving data-retrieval prototype.
- Migrated the network from a single node to a multi-node deployment for scalability and fault tolerance.
- Researched searchable encryption for private data retrieval.
- Hyperledger Fabric
- Blockchain
- Cryptography
- Distributed systems
Virtual Labs (VLEAD), IIIT Hyderabad
Hyderabad, India
Software Development Engineer Intern
–
- Built 10 interactive Digital VLSI simulations for Virtual Labs, a Government of India (MHRD) e-learning initiative.
- Modernized a 10-year-old legacy site, eliminating frequent hangs and high latency.
- JavaScript
- jsPlumb
- Bulma
- HTML
- CSS
Education & honors
Education & honors
IIIT Hyderabad
B.Tech with Honors in Computer Science and Engineering
Nov 2020 – May 2024 · Hyderabad, India
- CGPA 9.18/10 · Dean’s List (top 5%)
- Honors research in security at CSTAR Lab
- Teaching Assistant, Introduction to Information Security (Jan–Mar 2023): taught tutorials, created and graded assignments and exams
Coursework
- Data Structures and Algorithms
- Computer Networks
- Systems and Network Security
- Machine Learning
- Statistical Methods in AI
- Software Engineering
Honors & highlights
2nd place, Agentic AI track · JacHacks 2026
Ann Arbor. Won with ClaimSure AI, an agentic AI that helps people fight wrongful healthcare claim denials.
Speaker · a2Tech360 2026
Tech talk on ClaimSure AI at the Ann Arbor SPARK event.
Team Spotlight Award · Wells Fargo
For the multi-agent AI marketing-orchestration platform.
Dean’s List · IIIT Hyderabad
Top 5% (Spring 2023, Spring 2024); top 10% (Monsoon 2022, Monsoon 2023); Merit List, top 20% (2020, 2022).
JEE Main 2020 · All India Rank 841
Top 0.1% of 1M+ candidates. JEE Advanced 2020: AIR 2297.
Projects
Things I've built
Recent work on AI agents first, then earlier projects in ML, full-stack and systems.
- In progress · U-M
Declarative Computer-Use Interface for macOS Agents
Aug 2026 – Present
Porting the Declarative Model Interface (Wang et al., 2025: +67% task success, 43.5% fewer steps on Windows) to macOS. A graph of app menu states plus a path planner turns one declarative request into the full click sequence, cutting agent actions per target state from 4–5 to 1.
Advanced Scalable Systems, University of Michigan
- Python
- LLM agents
- Computer use
- macOS
GenAI Incident-Resolution Platform
Mar 2025 – Apr 2025
Full-stack GenAI platform for support engineers: a context-aware chatbot, automated root-cause summaries, related-incident retrieval and health-check generation, backed by vector similarity search over Gemini embeddings in ChromaDB.
- React
- Flask
- LangChain
- Gemini
- ChromaDB
- RAG
- 2nd place · JacHacks 2026
ClaimSure AI
Sep 2026
Agentic AI that helps people fight wrongful healthcare claim denials, with human-in-the-loop feedback on important decisions. Built in the Jac programming language at our team’s first hackathon at U-M.
Presented as a tech talk at a2Tech360 (Ann Arbor SPARK). Team: Abhayjit Singh Gulati, Adithyan Raja, Shankhanil Borthakur.
- Jac
- Agentic AI
- LLMs
- Human-in-the-loop
More projects
Unix Shell in C
Sep 2021 – Nov 2021
Unix-like shell with 8+ built-in commands, job control (fork, execvp, signals), I/O redirection, multi-level piping and /proc interaction.
- C
- Linux
- System calls
Canteen Ordering Portal
Jan 2022 – Feb 2022
MERN food-ordering app with customer and vendor portals, JWT role-based auth, real-time order tracking and email notifications. 3 services containerized with Docker Compose.
- React
- Node.js
- Express
- MongoDB
- Docker
Aspect-Based Sentiment Analysis
Sep 2022 – Dec 2022
Gated convolutional networks (GTRU) with aspect embeddings, implemented from scratch in PyTorch for ACSA and ATSA on 3 SemEval benchmarks (2014–2016), including hard mixed-sentiment subsets. Beat LSTM-with-attention baselines on speed and accuracy.
- PyTorch
- NLP
- Deep learning
Multi-Robot Minimum-Time Navigation with Deep RL
May 2024
Trained and compared DQN, DDPG, PPO and SAC to drive 2 robots to their goals in minimum time through a 10 m × 10 m arena with 7 obstacles, without access to the system dynamics, while keeping 0.2 m obstacle clearance and 0.5 m robot separation.
- Python
- DQN
- DDPG
- PPO
- SAC
Skills
Skills & certifications
Top skills
- Python
- Java
- Node.js
- AI Agents
- Google ADK
- LangChain
- Apache Spark
- Microsoft Azure
AI & ML
- AI agents / multi-agent systems
- LLMs
- Google ADK
- LangChain
- LangGraph
- RAG
- Model Context Protocol (MCP)
- A2A
- Embeddings & vector search (ChromaDB)
- LLM tracing & evaluation (Arize Phoenix)
- PyTorch
- NLP
- Deep reinforcement learning
Languages
- Python
- Java
- C++
- C
- JavaScript
- TypeScript
- SQL
- Bash
- Go (beginner)
Backend
- REST APIs
- Spring Boot
- Node.js
- Express
- NestJS
- Flask
- React
- Next.js
- Distributed systems
Data
- Apache Spark
- Spark SQL
- ETL / data pipelines
- Adobe Experience Platform
- PostgreSQL
- MySQL
- MongoDB
- SQLite
Cloud & DevOps
- Microsoft Azure (4x certified)
- AWS
- Docker
- Kubernetes
- Red Hat OpenShift
- Helm
- GitHub Actions
- Harness
- Jenkins
- Linux
- Git
Certifications
4× Microsoft Azure certified
Azure AI Engineer Associate
Microsoft
- Issued:
- Aug 2025
- Credential ID:
- 666788126FE53492
Azure Developer Associate
Microsoft
- Issued:
- Nov 2024
- Credential ID:
- 648C543C7982E692
Azure Administrator Associate
Microsoft
- Issued:
- May 2025
- Credential ID:
- 5BA72AA16051CD53
Azure Fundamentals
Microsoft
- Issued:
- Sep 2024
- Credential ID:
- EC0314F101476896
Deep Learning Specialization
DeepLearning.AI (Coursera)
- Issued:
- Apr 2025
- Credential ID:
- FPRXAM5GOK6R
Master Spring Boot 3 & Spring Framework 6 with Java
Udemy
- Issued:
- Sep 2024
DevOps Beginners to Advanced with Projects
Udemy
- Issued:
- Jun 2024
Contact
Get in touch
Hiring for Summer 2027 SWE or AI Engineering internships (US, May–Aug)? Want to collaborate, or just talk about my work? I'd love to hear from you.
Email me directly
Click the to reveal my address, then click it to write to me or copy it.
sri•••••• [at] gmail [dot] com