datafruit

AI DevOps Engineer

Summer 2025

Active

https://datafruit.dev/

AI DevOps Engineer

Datafruit is the first agentic DevOps engineer. It understands your deployment standards, design docs, and DevOps practices, then reasons across them to reduce cloud waste, prevent deployment delays, and eliminate wasted time across the stack. Reach out to us: founders@datafruit.dev

Active Founders

Venkat Arun

Founder

cofounder at datafruit - building the first ai dev ops engineer; previously cs and statistics at uc berkeley; worked on cloud infrastructure and data pipelines at various early stage startups.

Venkat Arun

Founder

cofounder at datafruit - building the first ai dev ops engineer; previously cs and statistics at uc berkeley; worked on cloud infrastructure and data pipelines at various early stage startups.

Abhiram Chennupati

Founder

Cofounder at datafruit, building the first agentic devops engineer. Previously worked on ML Infra at Amazon's search ranking team, did ML Research at the Space Sciences Lab, and majored in CS and City Planning at Berkeley

Abhiram Chennupati

Founder

Tom Jeong

Founder

Cofounder at datafruit, working on the first Agentic DevOps Engineer. Previously studied CS and theoretical mathematics and did quantitative risk analysis research at Georgia Tech. At Georgia Tech, I did research on using LayerSkip architecture in Mixture of Experts LLM model.

Tom Jeong

Founder

Nick Papciak

Founder

Co-founder of datafruit — the AI DevOps engineer. Previously worked at Meta. Studied machine learning and theoretical computer science at Georgia Tech, where I was the head TA for algorithms.

Nick Papciak

Founder

Co-founder of datafruit — the AI DevOps engineer. Previously worked at Meta. Studied machine learning and theoretical computer science at Georgia Tech, where I was the head TA for algorithms.

Company Launches

Datafruit: AI DevOps Engineer

See original launch post

Hi, we're Venkat, Abhi, Nick, and Tom — founders of Datafruit.

https://youtu.be/px6Gn4U3oPY

The problem: DevOps teams lose countless hours to operational interruptions. Managing temporary access, spinning up preview environments, rotating secrets, investigating cost spikes. Whether it's routine tasks that eat up time or complex issues requiring deep company knowledge - they all pull you away from strategic work

Our solution: Datafruit is an AI agent that learns your infrastructure, deployment standards, and design docs, then handles ops work through natural language requests in Slack or our UI.

Once Datafruit indexes your environment, you can ask it to do things like:

"Grant @Tom write access to analytics S3 bucket for 24 hours" Creates temporary IAM role, sends credentials, auto-revokes tomorrow.
"Create preview environment for PR #20" Spins up an isolated environment with test data, posts URL in PR, tears down on merge.
"Rotate API keys for x-service without downtime" Implements dual-secret rotation, updates deployments, verifies health, deprecates old keys.
"Why did Data Warehouse costs spike yesterday?" Identifies expensive queries, shows optimization options, implements fixes.
"Migrate our payment service from EC2 to EKS" Containerizes the app, creates EKS cluster with node groups, migrates volumes/configs, implements zero-downtime deployment, updates CI/CD pipelines.

Who we are: We are best friends from high school who have spent a ton of time causing (and fixing) vulnerabilities on projects we've made together. After high school, we went to UC Berkeley and Georgia Tech. While there, Abhi taught Computer Security at Berkeley, and Nick was the Head TA for algorithms at Georgia Tech. We then went on to work on cloud infra at different high-growth startups and Amazon, respectively.

Our ask: Book a call: cal.com/venkatl

Platform/DevOps leaders at 50-500 person companies: We can reduce your operational overhead.
Building a platform team? Skip months of internal tool development
Know DevOps/ Platform engineers facing these challenges? We'd appreciate an intro

YC Photos