02 AT RISK
Legacy .NET stack is holding the roadmap hostage
Every change costs a quarter. The modernization business case has been rewritten four times.
Technical transformation · Principal-led
Complex change.
Made operational.
16+ years turning fragile delivery into operating capability: Microsoft Azure and CAF-led cloud adoption, legacy modernization, live database migrations, and GenAI moved from demo to production governance. I lead every engagement personally and add specialists only where the gap is real.
0+Years
0+Programs
0%CSAT uplift
Transformation intelligence Scroll / resolve
- 01Assess
- 02Design
- 03Deliver
- 04Handover
AI-ready operating systems · 2026
[02/Problems I'm called for]
01 / Legacy systems
02 / Cloud & AI adoption
01 STALLED
Your AI pilot never became an operating capability
Copilot licences are live, a RAG demo impressed the board, and nothing behind it — data contracts, evaluation, ownership, funding — was ever designed.
03 AT RISK
Azure adoption without a landing framework
Subscriptions exist; governance, cost control and migration sequencing do not.
03 / Delivery complexity
04 CRITICAL
Delivery slips while every status stays green
Dates move quietly, and root cause sits between Engineering, QA, DevOps and Security — each reporting progress honestly.
05 CRITICAL
The team scaled faster than its operating model
Headcount tripled; decision-making did not change. Coordination now consumes the capacity you bought.
04 / Data & integration
Data & platform
- MySQL/MariaDB → PostgreSQL migrations
- Zero-downtime live data cutovers
- Elasticsearch, SQL Server, Redgate
- API-first integration design
[03/The operating model]
The engagement ends. The operating model keeps working.
An operating model is not a document. It is the set of decision rules, cadences and measurements that make good delivery repeatable without me in the room — designed against your constraints, installed on live work, handed to named owners.
- Owned by
- Your named leads
- Depends on me
- Never
- Installed on
- Live delivery
- Lifespan
- Beyond handover
01 / Assess
Instrument the program
Flow, decision latency, rework and handoff loss get measured before anything is designed. Those four numbers become the baseline and stay on the wall after handover.
Team size
Principal only — 1
Saurabh, embedded with your delivery and architecture leads
Deliverable
Problem statement, instrumented baseline, go / no-go recommendation
Note
Fixed fee. If the problem isn't mine to solve, I say so and stop here.
02 / Design
Design the operating model
Decision rights, cadence, intake and prioritisation rules, definition of done, escalation thresholds — written against Microsoft CAF where cloud is in scope, not against a methodology brochure.
Team size
Principal + 1 specialist
Architecture or data specialist added only where the assessment demands it
Deliverable
Target architecture, CAF-aligned roadmap, operating model v1
Note
The team expands here for the first time — and only by the size of the gap.
03 / Deliver
Install it on live delivery
No sandbox pilot. The model runs against real releases and real migrations while I am still in the room to correct it.
Team size
Principal + 2–4 specialists
Right-sized squad working on live migrations and releases beside your team
Deliverable
Model running in production, internal owners trained on real work
Note
Peak team size. Every specialist has a written exit condition from day one.
04 / Handover
Transfer ownership, then leave
Named internal owners, runbooks, and a KPI that fails loudly when the model drifts. The handover date is agreed in week one and it does not move.
Team size
Zero — model only
Transfer to named internal owners
Deliverable
Runbooks, drift KPI, 30/60/90 check-ins, signed handover
Note
The exit date is agreed in week one. It is a deliverable, not a negotiation.
Handover date fixed in week one
01
AI & agentic delivery
- Azure OpenAI, Copilot & GitHub Copilot rollout
- RAG architecture and evaluation loops
- Agentic workflows, n8n automation
- Prompt standards, MLOps fundamentals
02
Cloud & modernization
- Microsoft Cloud Adoption Framework (CAF)
- Azure architecture, DevOps, CI/CD
- Microservices, SAGA, Kafka, Debezium
- .NET → Node.js / NestJS revamps
03
Program leadership
- Technical program management at scale
- Agile delivery governance and KPIs
- Cross-functional: Eng, QA, DevOps, Security, CS
- Executive reporting and risk mitigation
04
Data & platform
- MySQL/MariaDB → PostgreSQL migrations
- Zero-downtime live data cutovers
- Elasticsearch, SQL Server, Redgate
- API-first integration design
[04/Proof of work]
Four problems seen, four tools shipped. Free, working, no signup.
Each started as friction inside a client program. Same instinct as the delivery work: find the real problem, build the smallest durable thing that removes it, hand it over.
Tool 01Shipped
SnapText
- Problem seen
- Analysts rebuilding decks by retyping PDF copy
- Tool built
- Browser-side text normaliser: strips hard breaks, hyphenation and page furniture
- Outcome
- Hours of retyping removed from every report cycle
Tool 02Shipped
GST Invoice Generator
- Problem seen
- Finance chasing invoice format errors every cycle
- Tool built
- Line-item invoice builder with CGST/SGST split and print-to-PDF
- Outcome
- Zero-rework invoicing for small finance teams
Tool 03Shipped
StatementIQ
- Problem seen
- Statement reconciliation done by eye, monthly
- Tool built
- Local CSV parser that buckets counterparties and flags recurring debits
- Outcome
- Reconciliation became a review, not a rebuild
Tool 04Shipped
PDF Converter
- Problem seen
- Scans arriving as loose image files across email threads
- Tool built
- Ordered image-to-PDF assembler, entirely on-device
- Outcome
- Document packs assembled without uploading client data
Interactive resume
Skills matrix, competence alignment and the full career timeline, explorable.
[05/How delivery runs]
Team size
Principal only — 1
Saurabh, embedded with your delivery and architecture leads
Deliverable
Problem statement, instrumented baseline, go / no-go recommendation
Note
Fixed fee. If the problem isn't mine to solve, I say so and stop here.
Team size
Principal + 1 specialist
Architecture or data specialist added only where the assessment demands it
Deliverable
Target architecture, CAF-aligned roadmap, operating model v1
Note
The team expands here for the first time — and only by the size of the gap.
Team size
Principal + 2–4 specialists
Right-sized squad working on live migrations and releases beside your team
Deliverable
Model running in production, internal owners trained on real work
Note
Peak team size. Every specialist has a written exit condition from day one.
Team size
Zero — model only
Transfer to named internal owners
Deliverable
Runbooks, drift KPI, 30/60/90 check-ins, signed handover
Note
The exit date is agreed in week one. It is a deliverable, not a negotiation.
Handover date fixed in week one
[06/Track record]
- 0+
- YEARS IN ENTERPRISE IT
- 0+
- PENSION TECH PROGRAMS
- 0%
- CLIENT SATISFACTION UPLIFT
Engineer to Associate Director
US retirement platforms, plus cloud and modernization work
Alongside a 60% gain in operational efficiency
Domains
- Aviation & MRO
- Pensions & Social Security
- Telecom
- Enterprise SaaS
- Lifestyle
LTIMindtree
2026 — PresentAssociate Director — Architecture / TPM
Azure cloud transformation and enterprise modernization for Microsoft CAF engagements. Cross-functional leadership across Engineering, QA, DevOps, Architecture, Security and Customer Success.
flydocs (Lufthansa Group)
2022 — 2025Technical Manager
End-to-end product lifecycle in airline IT. Led a full-stack revamp from legacy .NET to Node.js/NestJS and a live MySQL/MariaDB → PostgreSQL migration using Kafka and Debezium.
Azilen Technologies
2021 — 2022Tech Lead
Lifestyle-domain platforms for international clients including Hard Rock Café.
Sagitec Solutions
2014 — 2020Tech Lead
Enterprise systems for US pension, retirement and social security programs.
Tech Mahindra
2010 — 2014Sr. Technical Associate
Integrations and software support for global telecom clients including BT and AT&T.
Saurabh Churadiya
Associate Director — Architecture / TPM
Based in Indore, India; engaged with LTIMindtree, Pune. Leads Azure cloud adoption and modernization under Microsoft's Cloud Adoption Framework, with a delivery background spanning airline records systems, US pension platforms and global telecom integrations.
- Recognition
- ‘Great Manager’ survey
- Security risk
- −90% (US finance)
- Legacy processing
- −20% telecom & .NET Core
- Languages
- Hindi, English, Marathi +
Certifications
- Microsoft Azure Fundamentals
- Generative AI Fundamentals
- Agile Project Management Practitioner
- Practical Project Management
- Business Analytics with Power BI
- MS Office with Claude AI
[07/Engagements]
Priced on the outcome and the scope. Never on hours.
T-01
Assessment
Two weeks. One principal. A decision you can act on.
- +Instrumented baseline of the failing program
- +Root-cause problem statement, not a symptom list
- +AI / cloud readiness verdict: build, reframe, or don't spend
- +Fixed price, agreed before the start
Fixed-fee assessment
T-02Most engaged
Transformation
Architecture and operating model installed on live delivery.
- +Everything in Assessment
- +CAF-aligned Azure adoption or modernization roadmap
- +Right-sized specialist squad, scoped per phase
- +Priced on the outcome, not the calendar
Outcome-priced engagement
T-03
Handover
Ownership transferred, drift instrumented, team gone.
- +Runbooks and decision records for internal owners
- +Delivery KPIs that fail loudly on drift
- +30/60/90 review checkpoints
- +Signed exit with no retainer tail
Scope-based handover
[08/Start a conversation]
Describe the problem. I'll tell you if it's mine to solve.
Every engagement starts with a two-week assessment. If the assessment says the problem isn't mine to solve, I say so and stop there — that answer is worth the fee on its own.
Response within 2 business days