Building in public

Akshay Dixit

Senior Product Manager at Microsoft · Office Extensibility · 11+ years.
I build the tools I wish existed.

Akshay with Royal Enfield Himalayan at Wagamon Hill
Bruges canal
At Sela Pass
SCROLL

Most PMs check email first. My AI checks it for me.

I'm a Senior Product Manager at Microsoft with 11+ years in cloud engineering, developer platforms, and product strategy. Currently on the Office Extensibility team. Previously Azure Data Explorer (Kusto), where "kustonaut" was born.

I manage 6,000+ open GitHub issues across the Office Add-ins platform and have led teams of 15+ engineers. To handle the scale, I built Brain OS. It pulls signals from email, Teams, ADO, and GitHub every morning. Triages them. Generates a daily brief before my first coffee.

Microsoft Advocate. LinkedIn Top Voice in Product Management. INSEAD alumnus. Conference speaker. 365-day KQL challenge completer. Solo traveler when not shipping. I believe the best PM tools come from PMs who code.

Product Management AI Automation Developer Platforms Open Source Data Engineering Microsoft Advocate LinkedIn Top Voice Conference Speaker INSEAD Solo Travel
Akshay Dixit
Royal Enfield in Kerala Arunachal Pradesh Bruges at night 26000km across India Kanyakumari Feeding India volunteering Sunset Bruges panorama Royal Enfield in Kerala Arunachal Pradesh Bruges at night 26000km across India Kanyakumari Feeding India volunteering Sunset Bruges panorama

Impact in public view.

0
GitHub Issues Managed
OfficeDev ecosystem
0
Automations Built
Python · PowerShell · KQL
0
Copilot Skills Authored
Skill-as-Markdown
0
Stars on 365daysofADX
Community recognition

Three chapters. One thread: build what’s missing.

2014 — CHAPTER 1
Cloud Engineering & Operations
Microsoft · India
Started as a Program Manager building cloud infrastructure and supply chain systems. Learned to ship at scale, lead cross-functional teams, and develop a product mindset from day one.
Cloud Infrastructure Led 15+ Engineers
2022 — CHAPTER 2
Azure Data Explorer (Kusto) & Eventhouse in RTI Fabric
Microsoft · Azure Data
Became "kustonaut." Deep-dived into KQL, data engineering, and analytics at petabyte scale. Drove Eventhouse in Real-Time Intelligence for Microsoft Fabric. Completed a 365-day public KQL challenge. Built open-source tooling that earned 37+ GitHub stars.
365daysofADX ★ 37 KQL Cheat Sheet ★ 11 Eventhouse · RTI Fabric
2026 — CHAPTER 3
Office Extensibility & Developer Experience
Microsoft · India
Took ownership of developer reliability for the Office Add-ins ecosystem. Managing 6,000+ GitHub issues, cross-team partner programs, and the full developer support pipeline. Side project: building Brain OS — an open-source, AI-powered operating system for PMs with 27 Copilot skills and 50+ automations.
6,000+ Issues Developer Reliability Brain OS (Side Project) 27 Copilot Skills

Tools I built because nothing else worked.

Each project started from a real problem. Not a tutorial. Not a weekend hack. A genuine wall I hit at work.

kustonaut.github.io/issue-sentinel
Active

issue-sentinel

AI-powered GitHub issue triage. Classify, prioritize, and route issues automatically. Built for teams drowning in backlogs.

Python CLI GitHub Actions AI
Complete

365daysofADX ★ 37

A year-long deep dive into Azure Data Explorer. Daily KQL challenges, patterns, and learning resources. The project that started my data obsession.

KQL Azure Data Explorer Data Engineering
View on GitHub
kustonaut.github.io/kql-cheat-sheet
Complete

kql-cheat-sheet ★ 11

Interactive KQL reference guide. Born from my own need to stop Googling the same Kusto operators every week. Quick-lookup syntax for data engineers.

HTML KQL Reference
kustonaut.github.io/WealthPulse
Active

WealthPulse ★ 1

Self-hosted portfolio dashboard for Indian and US markets. 14 broker parsers, NPS tracker, EPFO integration, FIRE calculator, dark mode, live stock marquee.

Python Finance Dashboard
kustonaut.github.io/StockScreener
Stable

StockScreener ★ 2

Stock screening and analysis tool. Filter stocks with customizable criteria, track patterns, and make data-driven picks. Built for personal use, shared for everyone.

Python Data Analysis Stocks

What I work with daily.

🐍
Python
Primary
PowerShell
Automation
🔍
KQL
Data
🤖
GitHub Copilot
AI
🧠
GPT-4o / Claude
LLMs
📊
Azure Data Explorer
Analytics
🔗
Microsoft Graph
Platform
📋
Azure DevOps
Planning
💻
VS Code
Editor
🐙
GitHub Actions
CI/CD
📝
Markdown
Docs
🎨
HTML / CSS / JS
Web

Three rules I ship by.

01

Build what you need.

Not what looks good on a portfolio. Not what gets likes on Twitter. If I don't use it daily, I don't build it. Every repo here solves a problem I actually have.

02

Ship fast. Iterate faster.

My code ships before my slides. I'd rather deploy a rough v1 today than polish a perfect v1 next month. Done beats perfect. Every time.

03

Systems over goals.

I don't set targets. I build systems that make hitting targets inevitable. Brain OS isn't a productivity tool. It's a system that compounds intelligence daily.

I also write things that aren't code.

Travel stories, year-end reflections, and the occasional "how I balance Microsoft and solo trips" post.

Thought leadership & speaking.

Featured posts, conference talks, and milestones from a decade of building at Microsoft.

LinkedIn Top Voice — Product Management

What shaped how I think.

Books I've read cover to cover, docs I keep bookmarked, courses that actually changed how I work, and talks I rewatch.

Atomic Habits cover
Atomic Habits
James Clear
Rewired how I build systems. I stopped setting goals and started designing cue→routine→reward loops for everything—daily triage, spec writing, even exercise.
“1% daily compounds. Systems beat goals every time.”
Never Split the Difference cover
Never Split the Difference
Chris Voss
Changed every stakeholder negotiation I've had since. Tactical empathy, labeling, and calibrated questions work in product reviews, not just hostage situations.
“‘No’ is the start of negotiation, not the end.”
The Lean Startup cover
The Lean Startup
Eric Ries
Build→Measure→Learn isn’t just for startups. Every automation I ship follows this loop. Test with a script before building a pipeline.
“Vanity metrics lie. Ship the MVP. Learn fast.”
Deep Work cover
Deep Work
Cal Newport
I time-block 2–3 hours for spec writing with everything off. This book is why Brain OS handles shallow work so I can focus on what matters.
“The ability to focus without distraction is increasingly rare—and valuable.”
Extreme Ownership cover
Extreme Ownership
Jocko Willink & Leif Babin
When things go wrong, I say “this is mine.” No blaming engineering, no blaming process. The PM owns the outcome. Period.
“No bad teams, only bad leaders.”
Good to Great cover
Good to Great
Jim Collins
The flywheel concept runs my entire automation philosophy. Each improvement compounds. The Hedgehog Concept forced me to find my intersection of passion, skill, and impact.
“Flywheel: consistent turns beat heroic pushes.”
The Infinite Game cover
The Infinite Game
Simon Sinek
Stopped me from optimizing for quarterly metrics and started me building systems that outlast me. Worthy rivals, not competitors.
“Play the long game. Build things that outlast you.”
The E-Myth Revisited cover
The E-Myth Revisited
Michael E. Gerber
Work ON the system, not IN it. This is literally why Brain OS exists. If I’m always fighting fires, I’m not building systems that prevent them.
“Document everything as if you’re franchising.”

Let's build something.

PM tooling, AI automation, KQL deep-dives, or building things that shouldn't work but do. Always up for it.