Prashant Kumbhar — Senior Salesforce Solution Architect at Conversive. I turn tangled processes into clean, automated Salesforce architecture: Flows, Apex, and integrations that decision-makers can actually trust.
I've spent a decade inside the Salesforce ecosystem — starting as a Customer Success Engineer supporting clients across the US, and growing into the person clients now bring in to architect the org from the ground up. That path matters: I still think like the admin who has to maintain what gets built.
At Conversive, I lead projects that automate business processes end-to-end — collaborating with decision-makers to understand what they actually need, then designing it in Flow and Apex rather than duct-taping another workaround onto the org.
Alongside the day job, I hold an MSc in Corporate Management from the University of Europe for Applied Sciences (Berlin) and a BE in Computer Engineering from Savitribai Phule Pune University — a combination that keeps my solutions grounded in both the technical build and the business case behind it.
Every role below is tagged the way I'd tag an element on a Flow canvas — by the kind of work it actually was.
What I actually reach for when an org needs fixing.
Retiring Process Builder for good — record-triggered, screen, and scheduled Flows that hold up under real load.
Custom logic for the moments declarative tools can't reach — triggers, batch jobs, integrations.
Certified Sharing and Visibility Architect — designing security models that scale without fighting the users.
End-to-end configuration across the three clouds — objects, record types, layouts, and the automation between them.
Connecting Salesforce to AppExchange and third-party apps via API — without turning the org into a black box.
Sprint-based delivery inside fast-moving startups, shipping increments clients can react to quickly.
Real problems from inside live, high-volume Salesforce orgs — the kind that don't show up on a certification exam.
There was no structured sales process in the org, and leadership had no consolidated way to see pipeline health or team performance.
Designed and implemented a full sales process in Salesforce alongside leadership dashboards and reports, then connected the org to third-party lead-source APIs for automated lead capture, an SMS gateway for automated outbound messaging, and a cloud contact-center platform for computer-telephony integration.
Gave leadership real-time visibility into the pipeline and automated a chunk of the lead-to-outreach workflow that had previously been manual.
New customers needed hands-on help getting the company's SMS-automation product live inside Salesforce and Zoho, with no single owner for that process.
Took ownership of support, onboarding, and implementation for new customers — configuring Apex and Flow per customer environment, running proof-of-concepts, and mentoring newer team members as the process matured.
Built the documentation and playbooks that made onboarding repeatable, rather than something that only lived in one person's head.
A client needed custom asset and subscription management with no out-of-the-box Salesforce equivalent to lean on.
Owned the build solo — a substantial Apex codebase handling asset lifecycles, subscription terms, and renewals, layered under standard Flow-based automation for the rest of the process.
Delivered a working system that off-the-shelf tools couldn't cover, without a dedicated dev team behind it.
The org was hitting a data-storage ceiling, with legacy cases and tasks eating up space that newer, active records needed.
Designed and ran an archiving strategy for old cases and tasks to reclaim storage, then led a series of proof-of-concepts integrating our messaging product with Marketing Cloud, Marketing Cloud Next, Salesforce Sales Cadence, Microsoft Dynamics 365, and group messaging.
Freed up storage headroom and validated a roadmap of new integration surfaces for the product.
Every inbound case landed in one generic support queue — whether it came from a customer needing urgent help, or a client still in an early stage who'd benefit from one consistent point of contact.
An automated routing layer in Flow that splits cases in real time: urgent cases go through Omni-Channel to the next available support agent, while ongoing-relationship cases route to the customer's assigned Customer Success Manager. Unattended high-priority cases automatically escalate to a higher-priority queue after a set SLA window, so nothing sits untouched.
Customers got faster response times and a more consistent experience, while the support team stopped losing urgent cases in a shared queue.
Customer Success Managers were assigned to accounts manually — an approach that didn't scale and left senior agents overloaded while newer joiners sat underutilized.
A dynamic, formula-driven assignment engine that scores each CSM's current workload against their seniority-based capacity, then assigns new customers to whoever has the most headroom — with a built-in ramp-up curve so new hires aren't overwhelmed on day one.
Replaced a manual, spreadsheet-driven process with a self-balancing system that adjusts automatically as volume and team size change.
Salesforce data storage was nearly maxed out, with over a million email records tied to hundreds of thousands of cases consuming the majority of it — and storage costs in Salesforce scale far faster than in raw cloud storage.
A bulkified, self-throttling Apex pipeline (scheduler + queueable jobs) that identifies eligible closed cases, converts their emails to JSON, uploads them to an AWS S3 bucket, then replaces the original records with lightweight text-only copies for quick in-app reference — with built-in retry logic and failure alerting.
Freed up the majority of consumed storage without losing agent access to case history, at a fraction of the cost of scaling Salesforce storage directly.
Agents needed to make and receive calls without leaving Salesforce, and incoming calls needed to reach the right person — the assigned CSM if one existed, or an available support agent if not.
A Twilio integration inside Salesforce for initiating voice and video calls directly from a case or contact, paired with Apex logic that recognizes the caller, checks for an existing assigned CSM, and routes the call accordingly — falling back to standard queue routing when no CSM is found.
Gave agents a single workspace for calls and cases, and made sure returning customers consistently reached someone who already knew their history.
Live numbers from my Trailblazer profile.
Enterprise-scale data architecture & performance
Issued Jan 2026Sharing, security & data modeling mastery
Issued Jan 2026Configuring & optimizing Agentforce
Issued Apr 2025Customer-facing site implementation
Issued Apr 2024Secure, scalable sharing design
Issued Mar 2024Trusted AI strategy & principles
Issued Oct 2023Requirements gathering to solution design
Issued Apr 2023Core CRM platform fundamentals
Issued Sep 2022Scalable service solution design
Issued Aug 2020Sales solution design & delivery
Issued Jul 2020Declarative app development
Issued May 2020Apex & Visualforce development
Issued Jan 2019Advanced org administration
Issued Apr 2018Where the certification journey started
Issued Mar 2018Whether it's a messy migration off Process Builder, a security model that needs rethinking, or a full Salesforce build from zero — I'm happy to talk it through.