Swarag RavikumarI have spent roughly a decade working across healthcare technology, client-facing implementations, cloud initiatives and enterprise delivery. My focus is the operating discipline behind reliable execution: clear ownership, timely risk visibility, meaningful reporting and accountable decisions.
Healthcare integrations (2015–2019) → UK client-facing e-commerce, ERP and digital transformation (2020–2024) → portfolio-level governance and AI-enabled PMO work (2024–2026) → current cloud transformation responsibilities (2026–present).
Book a Delivery Health ConsultationCareer
From healthcare integrations to silicon and cloud portfolios — a track record built one delivered program at a time.
Project Manager
Bengaluru, India
Ongoing: global cloud migration and AI implementation for an enterprise client.
Project Manager
Bengaluru, India
Portfolio governance across cloud infrastructure, AI initiatives and silicon development; delivered an AI-enabled PMO.
Technical Account Manager
Bengaluru, India
Strategic relationships for enterprise accounts across CRM and marketing software.
Project Manager
London, United Kingdom
Digital transformation programs for growing UK businesses with teams of 15-20.
Project Delivery Specialist
Nottingham, United Kingdom
Global ERP implementations for rental and asset management customers.
Customer Success Specialist
Leicester, United Kingdom
Customer success initiatives for global e-commerce brands.
Junior Project Manager
Leicester, United Kingdom
Delivered e-commerce and events projects with a five-person team.
Project Delivery Specialist
Bengaluru, India
Delivered healthcare integration projects for GP practices, insurers and laboratories.
Trainee Project Specialist
Bengaluru, India
U.S. healthcare accounts across GP offices, laboratories and insurance.
A practical approach
Understand delivery pain, risk, governance and decision bottlenecks.
Artifact · Project health assessment
Identify critical problems and agree measurable success criteria.
Artifact · Risk-ranked action plan
Introduce appropriately scoped controls, reports and workflows.
Artifact · Governance cadence + dashboard
Review results, refine practices and maintain accountability.
Artifact · Monthly executive action report
Signature engagements
Selected responsibilities, described by role and industry. Client and employer details are kept confidential.
Core expertise
Delivery & workflow platforms
ORQYNETechnical knowledge
AWS Cloud FundamentalsCredentials
Nottingham Trent University
MSc Engineering Management - Distinction
Nottingham, United Kingdom • 2019 - 2021
Global Academy of Technology
BE Mechanical Engineering - Distinction
Bengaluru, India • 2011 - 2015
AI stack & model playbook
I designed and delivered an AI-enabled PMO ecosystem, and I currently run an AI rollout inside a governed cloud migration plan. Below: the tools I build with, and an honest view of which model belongs on which job.
ORQYNENew launch
Next-generation project and business workflow management platform - orchestrating plans, approvals and operational workflows in one governed system.
Reasoning & analysis
Program strategy, risk analysis, stakeholder comms and executive narrative drafting.
Long documents
Contract, scope and specification review with long-context reading and careful structured output.
Multimodal
Slide decks, screenshots, video and spreadsheet inputs turned into structured delivery artefacts.
Code generation
Automation scripts, integrations and PMO tooling built from plain-language requirements.
Agentic builds
Multi-file builds and refactors run as an agent inside a real repository.
Agentic app builds
Rapid end-to-end app generation for internal tools and proof-of-concepts.
Product prototyping
Production-ready web apps and portals shipped in days, not sprints.
Choosing the wrong model burns budget and trust. This is the selection logic I apply inside delivery teams.
Complex planning, risk and trade-off analysis
GPT-6.1 reasoning
Strong multi-step reasoning and planning.
Reasoning
Context
Speed
Value
Claude Opus 5.5
Deep analysis, higher nuance.
Reasoning
Context
Speed
Value
Gemini Pro-class
Large context, multimodal reasoning.
Reasoning
Context
Speed
Value
Why this works
Deep reasoning models hold many constraints at once - ideal for schedule modelling, dependency mapping and mitigation strategy.
Long document and contract review
Claude Opus 5.5 / Sonnet 4.6
Precise reading of long, dense documents.
Reasoning
Context
Speed
Value
Gemini Pro (long context)
Million-token context in one pass.
Reasoning
Context
Speed
Value
Kimi K3.1 Pro
Open-weight, 1M-token context.
Reasoning
Context
Speed
Value
DeepSeek V4
Open-weight, strong value at 1M context.
Reasoning
Context
Speed
Value
Why this works
Million-token context windows read an entire SOW, RFP or migration runbook in one pass without losing detail.
Building and refactoring software
Claude Code
Agentic coding across whole repositories.
Reasoning
Context
Speed
Value
Codex
Cloud coding agent that runs and tests code.
Reasoning
Context
Speed
Value
GPT-6.1 Codex-class
Coding-tuned model for refactors and reviews.
Reasoning
Context
Speed
Value
Why this works
Coding agents plan across files, run the code and self-correct - the right tool for automation and internal tooling.
High-volume summarising, status and classification
GPT-6 mini-class
Fast, low-cost everyday automation.
Reasoning
Context
Speed
Value
Gemini Flash
Very fast and cheap at scale.
Reasoning
Context
Speed
Value
Claude Haiku
Quick, reliable classification and summaries.
Reasoning
Context
Speed
Value
DeepSeek V4 Flash
Ultra-low cost high-volume processing.
Reasoning
Context
Speed
Value
Why this works
Fast, cheap models handle repetitive PMO work - meeting notes, status roll-ups, ticket triage - at a fraction of the cost.
Research and synthesis across sources
Deep research modes
Agentic browsing with cited sources.
Reasoning
Context
Speed
Value
Gemini Pro with grounding
Answers grounded in live search.
Reasoning
Context
Speed
Value
Why this works
Tool-using models gather, verify and cite sources instead of guessing - essential for vendor and technology evaluations.
Multimodal inputs: diagrams, screenshots, recordings
Gemini Omni
Native video, audio and image understanding.
Reasoning
Context
Speed
Value
GPT vision-capable models
Reads diagrams, screenshots and dashboards.
Reasoning
Context
Speed
Value
GPT Image 2.6
Generates and edits visuals, diagrams and SVG.
Reasoning
Context
Speed
Value
Why this works
Architecture diagrams, dashboards and call recordings become structured, actionable delivery data.
Enterprise search and work-context AI
Atlassian Rovo
Grounded in Jira and Confluence.
Reasoning
Context
Speed
Value
Microsoft Copilot
Grounded in Microsoft 365 work data.
Reasoning
Context
Speed
Value
Why this works
Models grounded in Jira, Confluence and M365 answer from your own governed program data.
Open-weight and private hosting
Kimi K3.1 Pro
Open-weight, 1M-token context.
Reasoning
Context
Speed
Value
DeepSeek V4
Open-weight, strong value at 1M context.
Reasoning
Context
Speed
Value
Mistral Large 4
European open-weight, self-hostable.
Reasoning
Context
Speed
Value
Why this works
Open-weight models run inside your own cloud or data centre - the right call when client data, IP or regulation means nothing can leave your environment.
Complex planning selected
Powered by AI, using the same selection logic above. Recommendations are guidance, not guarantees.