Piyush Ranjan

Piyush Ranjan

Technical Lead @ Barclays Corporate & Investment Bank

LinkedIn creator · United States · 32.2K followers
Open to collaborations

Companies pay Piyush to write a post about their product on LinkedIn.

This page is where they book it. Below: Piyush's audience, recent posts and price. Naano, the platform behind this page, handles the brief, your approval of the post and the payment.

€363per post, fixed price

or 3 posts for €1,000 · You approve the post before it goes live

Book Piyush

Looking for Piyush? View LinkedIn profileA creator yourself? Create your own page

About Piyush Ranjan

AI & Technology Leader | AI Architect | Generative AI, Agentic AI, FinTech, Fraud Detection | 32K+ LinkedIn Followers | Author | Patent Holder

I'm an AI and technology leader with 18+ years of experience across Artificial Intelligence, financial technology, enterprise architecture, cloud transformation, and digital innovation. I combine enterprise technology expertise with hands-on AI innovation, exploring how Generative AI, AI agents, machine learning, and intelligent automation can solve real-world business challenges. With a community of 35K+ followers on LinkedIn, I share practical insights, technology breakdowns, industry trends, and thought leadership content that makes complex technologies accessible to professionals, developers, architects, and business leaders. I'm also an author, German utility patent holder in AI and fraud detection, IEEE Senior Member, technology speaker, mentor, and industry contributor. What I focus on: → Generative AI & LLMs: Large Language Models, RAG, prompt engineering, enterprise AI applications, and AI-powered solutions. → Agentic AI & Multi-Agent Systems: Intelligent agents, agent orchestration, MCP, Google ADK, AI automation, and autonomous workflows. → AI in Financial Services: Fraud detection, financial risk management, intelligent surveillance, regulatory compliance, and financial security. → Cloud & Enterprise Architecture: AWS, Google Cloud, cloud-native applications, distributed systems, microservices, and scalable enterprise platforms. → Cybersecurity & Data Engineering: AI-driven security, data protection, API security, intelligent threat detection, and data-driven decision-making. → Emerging Technology & Innovation: Robotics, automation, developer tools, startups, digital transformation, and the future of technology. Content & Brand Collaborations I collaborate with technology companies, AI startups, SaaS platforms, and innovative brands to create authentic, educational, and engaging conten

SoftwareAIDeveloper Tools

My LinkedIn audience

Followers
32,174
Avg impressions/post
8.6K
Top 20 exported posts · median 6.4K
Engagement
2.6%

LinkedIn export · Nov 14, 2025 – Nov 13, 2026 · Provided by the creator, not independently verified by Naano

  • There is a massive gap between writing clever prompts and understanding LLM architecture. For engineers looking to bridge that gap, this Stanford series offers a structured, ground-up curriculum covering everything from attention mechanisms to autonomous agents. The 9-Lecture Curriculum (in order): Part 1: Transformer Fundamentals — https://lnkd.in/gaaRDexT Part 2: Transformer Models + Practical Tricks — https://lnkd.in/gy4FUwNY Part 3: Transformers to Large Language Models — https://lnkd.in/gsPiCrEU Part 4: How LLMs Are Trained — https://lnkd.in/gvHJvgqP Part 5: Tuning & Adaptation — https://lnkd.in/g6kgtPKR Part 6: Reasoning in LLMs — https://lnkd.in/gAACSUG6 Part 7: Agentic LLMs (Tools, Planning, Workflows) — https://lnkd.in/gVm6js9z Part 8: Evaluation: What "Good" Really Means — https://lnkd.in/gJhbFQ4s Part 9: Recap + What’s Trending Now — https://lnkd.in/g5JMNTsf Execution strategy: Treat each video as a module. Watch one, document the core mechanics, and run a quick local experiment to validate what you learned before advancing. A strong foundational resource to anchor your 2027 technical learning goals.

    166 reactions · 4 commentsView
  • One agent was holding up my entire AI workflow. Here's how I fixed it without scaling everything. I've been building multi-agent Python systems for a while — one agent handles ingestion, another does analysis, another formats output. The problem? One agent always becomes the bottleneck. Work queues up behind it while the others sit idle. The obvious fix is to scale the whole app — but that wastes resources on agents that don't need them. I recently tried CanyonOS on one of my workflows. The approach is different. Multi-agent Python workflows rarely have even load. One agent often ends up with most of the calls while the others sit idle. The obvious fix is scaling the whole app, which spends resources on agents that don't need them. I tried CanyonOS on a small 4-agent workflow I built: a research-paper title reviewer. → TopicAgent tags every title (plain Python) → ReviewAgent reviews each title with Gemini Flash-Lite, one call per title, in parallel → ReportAgent ranks the weakest titles first (plain Python) → A plain-Python orchestrator calls them like normal functions: tags = TopicAgent().tag(titles=titles) reviews = [ReviewAgent().review(title=t) for t in titles] CanyonOS runs each agent as its own service and scales each one separately on its own load, within the replica limits I set (1–3 per agent). What I saw in my run (166 requests, 830 model calls, 7 minutes, local Docker): 1. Only the busy agent scaled ReviewAgent went 1 → 2 → 3 replicas within the first minute and held at my limit of 3. The other three agents stayed at 1 the whole time. When the load stopped, it drained back to 1. Once integrated, CanyonOS added replicas without further code changes. The integration I did need: declaring each agent, a scaling policy and a model policy in CanyonOS's YAML config. 2. Request traces without tracing code I opened one request and followed all 8 agent calls: TopicAgent, 5 parallel ReviewAgent calls and ReportAgent, 905 ms end to end. No tracing decorators or instrumentation SDK in my agents. 3. Estimated model cost by agent That request: 539 tokens and about $0.00064 in estimated model cost, all attributed to ReviewAgent. The whole run came to about $0.10 in estimated model cost, 100% ReviewAgent. These are estimates from list prices, not my provider bill, and compute is separate. One lesson: my first try sent bursty traffic and only scaled sometimes. CanyonOS adds a replica after 4 straight readings above the threshold, and steady traffic made it consistent. Caveat: a local, small-scale run on CanyonOS 0.2.1. I'm not claiming speedups or savings, because I didn't measure them. If you're building Python multi-agent workflows with uneven load, check it out on Product Hunt 👉 https://lnkd.in/ea57JjZx Sponsored / Paid partnership with Canyon. All demo results are my own. #AI #Python #MultiAgent #LLM #AIEngineering #CanyonOS

    129 reactions · 13 commentsView
  • Multi-Agent Autonomous Delivery Pipeline Building fully autonomous software delivery pipelines is no longer science fiction. By chaining specialized AI agents together, we can bridge the gap between product requirements and production deployment with unprecedented velocity. I’ve mapped out the end-to-end architecture and orchestration flow for a Multi-Agent Autonomous Delivery Pipeline. Here is how the workflow runs: Jira Ingestion: A webhook captures new requirements and converts them into structured task graphs. Autonomous Coding: The developer agent clones the repo, writes code, utilizes RAG, and opens a Pull Request. Self-Correction QA Loop: The QA agent analyzes test failures, commits bug fixes, and re-runs CI until all tests pass green. UAT Deployment: The DevOps agent packages the container and rolls it out to staging, updating Jira automatically. Human-in-the-Loop Gate: Stakeholders review the feature in UAT and sign off with a single click to trigger zero-downtime production deployment. I've attached the architectural guide and orchestration code snippets above. How comfortable is your engineering team with letting AI agents push code directly to staging today? Let me know below! 👇 #AIgents #SoftwareEngineering #DevOps #Automation #Kubernetes #LLM #TechArchitecture

    7 reactions · 2 commentsView

Work with Piyush Ranjan

Choose a format for your next LinkedIn collaboration.

Sponsored LinkedIn post

A dedicated post to introduce your brand to my audience.

€363

3 LinkedIn posts

A series of posts to build visibility over time.

€1,000

LinkedIn repost

Reposts your LinkedIn post to my network.

€238

Repost with comment

Reposts your LinkedIn post with my own comment.

€238
Book Piyush

Your collaboration, managed on Naano

  • Share your brief with the creator
  • Review the content before publication
  • Manage your collaboration and payment
Create your brand account to continue. You can review the collaboration details before committing.

How does this price compare? B2B creator rates 2026

Who sees my posts

People who saw my posts, from my LinkedIn export.

Software Engineer10%
Founder2%
Data Engineer1%
Solutions Architect1%
Project Manager1%

Past partnerships

Storefront by Naano · Create yours

Booking Piyush Ranjan: frequently asked questions

Piyush Ranjan is a B2B LinkedIn creator based in the United States with 32,174 followers, posting about Software, AI, and Developer Tools. On Naano, one sponsored LinkedIn post by Piyush Ranjan costs €363, a fixed price per post.

How much does Piyush Ranjan charge for a sponsored LinkedIn post?
One sponsored LinkedIn post with Piyush Ranjan costs €363 on Naano. It is a fixed price per post, the amount your brand pays. Packs of several posts are listed in the offers above. Compare with the B2B creator rates 2026 report
How do I book Piyush Ranjan?
Create a brand account on Naano, choose an offer on this page and send your brief. The price is held when Piyush Ranjan accepts and paid when the post URL is submitted. You can turn on draft approval to review the post before it goes live.
Can Naano run the campaign for me?
Yes. With Managed Campaigns, the Naano team sets the strategy, casts the creators, writes the briefs, reviews the drafts, pays the creators and reports results per creator and campaign. It is a three-month programme: a management fee from €3,000 per month, plus a creator budget from €6,000 per month. See Managed Campaigns

Want Naano to run the whole campaign?

The Naano team casts the creators, writes the briefs, reviews every draft with you and reports results per creator. You approve the shortlist and every draft.

See Managed Campaigns
€363per postBook Piyush