AI · Architecture · Clarity

Clarity is a quiet
kind of power.

I'm Saurav Bakshi. I write The Thinking Hat — for the architects now expected to have the AI answer no one wrote down for them. I turn the noise into what actually changes a decision you own.

Read the essaysWork with me

Latest writing

From the newsletter

All essays
  1. 01Unlocking SaaS Potential — Build a Strategy That Works For YouSaaS only pays off if you change how you think. A strategy built on three pillars — business model, scalability, and integration — to unlock real ROI from your SaaS partners.29 Nov 2024
  2. 02The Hidden Dangers of Bespoke Data Integration PatternsCustom, one-off data integrations feel like a quick win — and quietly become a downward spiral for enterprise agility. The case for loose coupling and a canonical model.23 Nov 2024
  3. 03The Road to TOGAF CertificationHow I earned the TOGAF certification as a busy professional — the strategy, the resources, and the habits that made two focused months enough.11 Jul 2024

What I think about

Enterprise AI isn't short on information. It's short on information written for you — leaving the architect to turn the noise into a decision, alone.
01

Cut the noise

The engineer has docs, the executive has decks, the vendor has a pitch — none of it pitched where you actually decide. I strip the churn down to what truly touches the architecture.

02

Find the decision

Most AI writing explains what something is. The only question that matters is what it changes — when a choice is forced, what tradeoff appears, what breaks if you get it wrong.

03

Make it navigable

Complexity isn’t avoided; it’s structured. I turn a moving, uncertain field into something an architect can hold, explain, and act on — drawn from what I build, not from slideware.

Advisory

How I help

I partner with teams designing enterprise AI architecture — the systems, decisions, and tradeoffs of building on AI, designed for where things are heading, not where they have been.

AI-native architecture

Designing systems around intelligent, agentic services — where decisioning, automation, and orchestration are model-driven by default rather than bolted onto legacy flows.

  • Agentic service design
  • Model-driven decisioning
  • Event-driven orchestration

Cloud & integration

Loosely coupled, cloud-native foundations and clean integration patterns — so the right capabilities plug in without the usual technical debt and bottlenecks.

  • Cloud-native design
  • Loose coupling & APIs
  • Canonical data models

AI-ready foundations

Re-architecting the core stack so it can actually run on AI — clean data contracts, a decisioning layer, and the governance to deploy models safely inside a regulated enterprise.

  • Data & decisioning layers
  • Core-to-AI integration
  • Model governance & risk

Building on AI — and want a partner who's done the thinking?

Start a conversation

Get in touch

Let's talk

Working on a cloud, AI, or architecture challenge — or just want to compare notes? Drop me a line.

Send message
Elsewhere
Based
Remote · global advisory