Shubham Anandani

Engineering manager writing practical notes on Docker, Intel OpenVINO, Azure, Java, GitHub and Oracle tech, plus lessons from leading engineering teams. Originally published at imshubh.com.

Seattle, WA, US Joined on Oct 10, 2026 https://imshubh.com

Education

Masters in Software Engineering, Carnegie Mellon University

Work

Software Engineering Manager, LinkedIn

Shubham Anandani

Oct 10

Revocation Is the Hard Part

#identity #security #privacy #microsoft

3 min read

Want to connect with Shubham Anandani?

Create an account to connect with Shubham Anandani. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

Shubham Anandani

Oct 10

Designing for Partial Data: Distinguishing “We Don’t Know” From “False”

#java #programming #architecture #softwaredevelopment

3 min read

Shubham Anandani

Oct 11

Packaging Inference for Production: A Checklist for Shipping an Optimized-Model Service

#mlops #ai #devops #intel

4 min read

Shubham Anandani

Oct 11

Designing for Other People’s Code: A Plugin API That Welcomes Contributions

#opensource #github #api #architecture

3 min read

Shubham Anandani

Oct 11

Reproducible Java Builds: Pinning, Locking, and Clean-Room Verification

#java #devops #cicd #buildsystems

3 min read

Shubham Anandani

Oct 11

Keys Don’t Belong in the Image: Secret Management for Credential Signing

#security #azure #identity #devsecops

3 min read

Shubham Anandani

Oct 11

Deploying Identity Services Without a Server to Babysit

#azure #serverless #identity #containers

3 min read

Shubham Anandani

Oct 11

Null Isn’t False: Building Fair Credential Verifiers

#identity #security #microsoft #architecture

3 min read

Shubham Anandani

Oct 11

Docker in CI: The Real Cost of Docker-in-Docker

#docker #cicd #devops #testing

3 min read

Shubham Anandani

Oct 11

Flaky Tests Are a Concurrency Bug: Fixing Shared State Under Parallel Execution

#java #testing #concurrency #softwareengineering

3 min read

Shubham Anandani

Oct 11

CPU-Only, By Design: Building an Edge-AI Service Without Assuming Accelerators

#edgeai #ai #cpu #architecture

4 min read

Shubham Anandani

Oct 11

Virtual Threads Changed My Mental Model: Practical Lessons From I/O Fan-Out

#java #concurrency #performance #programming

4 min read

Shubham Anandani

Oct 11

Verifiable Doesn’t Mean Verified: Getting Issuer, Holder, and Verifier Right

#identity #security #microsoft #verifiablecredentials

4 min read

Shubham Anandani

Oct 11

Triage Without Burning Out: Sustainable Patterns for Issues and PRs

#opensource #github #maintainers #productivity

3 min read

Shubham Anandani

Oct 11

Don’t Ship What You Haven’t Scanned: Gating Images Before Production

#docker #security #devsecops #supplychain

3 min read

Shubham Anandani

Oct 11

Semantic Versioning Is a Promise, Not a Formality

#opensource #github #semver #devops

3 min read

Shubham Anandani

Oct 11

Records Aren’t Just Less Typing: Modeling Ambiguity Without Nulls

#java #programming #architecture #softwaredevelopment

4 min read

Shubham Anandani

Oct 11

A Breaking Change in Someone Else’s PR: Guarding Extension Points Without Blocking Contributions

#opensource #github #api #softwareengineering

4 min read

Shubham Anandani

Oct 11

Benchmarking Honestly: Building a Repeatable, Trustworthy Inference Benchmark

#benchmarking #ai #performance #intel

4 min read

Shubham Anandani

Oct 11

Contribution Guidelines Nobody Reads: Designing Onboarding That Actually Works

#opensource #github #onboarding #devrel

4 min read

Shubham Anandani

Oct 11

Smaller, Quieter, Safer: A Production Image Checklist

#docker #security #devops #containers

4 min read

Shubham Anandani

Oct 11

The Conversion Gap: From ONNX Export to an Optimized Runtime Format

#onnx #machinelearning #ai #intel

4 min read

Shubham Anandani

Oct 11

Faster Isn’t Free: The Real Trade-offs of Quantizing a Model from FP32 to INT8

#ai #machinelearning #performance #intel

4 min read

Shubham Anandani

Oct 11

Same Dockerfile, Different Build: Reproducible Container Builds

#docker #devops #containers #buildsystems

3 min read

Shubham Anandani

Oct 11

Running Is Not Ready: Operating Containerized Services

#docker #devops #kubernetes #sre

4 min read

DEV Community

Shubham Anandani

GitHub Repositories

Skills/Languages

Currently learning

Currently hacking on

Available for

Revocation Is the Hard Part

Want to connect with Shubham Anandani?

Designing for Partial Data: Distinguishing “We Don’t Know” From “False”

Packaging Inference for Production: A Checklist for Shipping an Optimized-Model Service

Designing for Other People’s Code: A Plugin API That Welcomes Contributions

Reproducible Java Builds: Pinning, Locking, and Clean-Room Verification

Keys Don’t Belong in the Image: Secret Management for Credential Signing

Deploying Identity Services Without a Server to Babysit

Null Isn’t False: Building Fair Credential Verifiers

Docker in CI: The Real Cost of Docker-in-Docker

Flaky Tests Are a Concurrency Bug: Fixing Shared State Under Parallel Execution

CPU-Only, By Design: Building an Edge-AI Service Without Assuming Accelerators

Virtual Threads Changed My Mental Model: Practical Lessons From I/O Fan-Out

Verifiable Doesn’t Mean Verified: Getting Issuer, Holder, and Verifier Right

Triage Without Burning Out: Sustainable Patterns for Issues and PRs

Don’t Ship What You Haven’t Scanned: Gating Images Before Production

Semantic Versioning Is a Promise, Not a Formality

Records Aren’t Just Less Typing: Modeling Ambiguity Without Nulls

A Breaking Change in Someone Else’s PR: Guarding Extension Points Without Blocking Contributions

Benchmarking Honestly: Building a Repeatable, Trustworthy Inference Benchmark

Contribution Guidelines Nobody Reads: Designing Onboarding That Actually Works

Smaller, Quieter, Safer: A Production Image Checklist

The Conversion Gap: From ONNX Export to an Optimized Runtime Format

Faster Isn’t Free: The Real Trade-offs of Quantizing a Model from FP32 to INT8

Same Dockerfile, Different Build: Reproducible Container Builds

Running Is Not Ready: Operating Containerized Services