Krishna Shukla

CLOUD.AI.AGENTIC.

Building cloud, AI, and data systems while figuring out how they actually work.

FIND ME

// HOW I'M GROWING

Engineering Trajectory

Where I started, what I'm building now, what I'm deliberately learning next, and where this is going — a progression through harder systems, not a five-year plan. Click any stage.

WHERE I AM NOW

FinOps Hub

An agent that turns plain-English cloud-cost questions into validated SQL over Databricks. The problem I'm chasing: let anyone interrogate spend without writing SQL.

FastAPIAzure OpenAIDatabricksNext.js
WHERE I AM NOW

Cloud infrastructure

Provisioning Azure and AWS from Terraform — hub-spoke networks, private endpoints, everything reproducible from code.

TerraformAzureAWS
WHERE I AM NOW

Data platforms

A bronze–silver–gold Databricks warehouse feeding cost attribution, anomaly detection, and reporting.

DatabricksDelta LakeADF

// SELECTED WORK

Selected work

A few projects I've built. Each write-up covers the problem, how it was built, and the result.

AI · Cloud · Data01
plain English interfaceread-only sqldeterministic answers

FinOps Hub — Autonomous FinOps Analyst

Ask your cloud-cost data a question in plain English. FinOps Hub turns it into validated SQL over a Databricks warehouse and returns a summary, charts, and the exact steps it ran.

Python / FastAPIAzure OpenAI (gpt-5-mini)Databricks SQLNext.js
Read case study
Cloud · Platform · DevOps02
~21 modules4 · peered vnetsprivate endpoints

Azure Hub-Spoke Landing Zone

A modular Terraform landing zone for Azure: a hub-spoke network with a central firewall, private DNS, and four peered VNets where every data and AI service sits behind a private endpoint.

TerraformAzure FirewallPrivate DNSVNet Peering
Read case study
AI · Cloud · Platform03
OAuth 2.0 authMCP protocolJWT-verified tokens

AWS Observability MCP Server

A Model Context Protocol server that gives AI agents safe access to AWS CloudWatch, secured with Cognito OAuth 2.0 and JWT validation.

FastMCPAmazon CognitoJWTCloudWatch
Read case study

// STACK

The stack I work in

Four areas I work across. Cloud carries delivery, delivery moves data, and data feeds the AI on top of it.

A
AI & GenAI
INTELLIGENCE
  • LangChain
  • LangGraph
  • Azure OpenAI
  • MCP
  • Prompt Engineering
C
Cloud
PLATFORMS
  • Microsoft Azure
  • AWS
D
DevOps & IaC
DELIVERY
  • Terraform
  • Azure DevOps
  • CI/CD
  • Docker
  • Kubernetes
D
Data
PIPELINES
  • Azure Data Factory
  • Databricks SQL
  • Pandas
  • Power BI
CloudDevOpsDataAI

// HOW I WORK

How I approach engineering

K
Krishna Shukla
AI & Cloud Engineer
8.8
B.Tech CGPA
3
Clouds
Agentic
AI Focus

How does this actually work?

I want to know how a system behaves underneath before I trust it, so I dig until the black boxes aren't black anymore. Most of my projects start as that one question.

Build it, break it, fix it

I learn a thing by building it, breaking it, and fixing it. Implementation over theory, every time — tutorials never taught me as much as a system falling over did.

Infrastructure is a product

I provision with Terraform, ship through CI/CD, and keep it observable, so the next person can read it and build on it instead of guessing how it runs.

// EXPERIENCE

Where I've built

Junior Associate — AI & Cloud Engineer

NOV 2025 — PRESENT

Celebal Technologies

Building FinOps Hub, a natural-language FinOps analyst on Azure and Databricks. It includes a hybrid template and LLM SQL engine, a FastAPI and Next.js streaming interface, a bronze-silver-gold warehouse, and an AWS observability MCP server, provisioned with Terraform and Azure DevOps.

Software Engineer

JUL 2025 — OCT 2025

Encodency Pvt Ltd

Built backend services and software, working across teams on application development and deployment, with version control, debugging, testing, and delivery as part of the routine.

// CONTACT

Get in touch

I'm open to AI and cloud engineering roles, and to interesting systems work. Email is the fastest way to reach me.