Saturday, April 18, 2026/Vol. I · No. 1

Sreekanth Payyavula

A Portfolio · Vol. I · GenAI × ERP

Engineering AI into the systems that already run the world.

By Sreekanth PayyavulaSenior Engineering Manager, Walmart Global TechSan Jose, CAUpdated April 2026

This is a working index of the software I have built, the systems I have rescued, and the ideas I keep coming back to. By day I lead engineering at Walmart Global Tech on GenAI and ERP modernization; independently I run SAALR, a studio on the same frontier for teams that do not have a Walmart to do it for them. The bias across both surfaces is for software that outlives its first launch — typed interfaces, honest errors, and AI that earns its keep by disappearing into a workflow rather than announcing itself.

Below you will find eight projects organized from pre-launch products down to open-source research, followed by half a dozen essays on the disciplines that make the shipping work possible. Some are written for the next engineer who inherits the system; some are written for the next client who wants to know how I think. None of them are launch copy.

§ 01 — Recent WorkUpdated quarterly

Where generative AI meets the enterprise backbone.

Featured · Studio · Founder

SAALR — GenAI, meet the ERP backbone.

An independent studio and toolkit for bringing generative AI to the ERP systems that actually run the global supply chain. Migration accelerators, agent-assisted customization, and a frank take on what “enterprise AI” means when the data is forty years old and the business has been running on it the whole time.

The studio was started after watching the same pattern play out in three separate enterprise programs: the vendor demo works on clean schema, the pilot works on a blessed slice of production, and the rollout dies quietly when it meets the full graph of customizations, permissions, and forty years of accumulated exceptions. SAALR starts from the full graph.

Filed · GenAI · ERPFounder · 2024—Visit saalr.io →
View the complete index →
§ 02 — From the Desk04 entries on file

Notes from the draft folder.

  • essayFeb 14, 2026
    Why ERP modernization is the real AI opportunity.

    The most boring systems in the enterprise are the ones agents are best positioned to change. A brief on why the ERP backbone — SAP, Oracle, the forty-year-old schemas that run purchase orders and payroll — is where GenAI will actually earn its keep, and why the shiny copilot demos will not.

    8 min
  • essayNov 03, 2025
    Notes from six months leading GenAI at enterprise scale.

    A balance sheet after two quarters embedding AI across internal platforms at Walmart Global Tech. Where it works, where it does not, which models made it into production, and the operating model that kept the whole program honest without starving the rest of the engineering org.

    12 min
  • noteAug 22, 2025
    Typed interfaces, honest errors.

    A field guide to error propagation that respects the caller: typed failure modes, bounded retries, and a hard rejection of the fallback that silently hides a broken system. The examples are TypeScript and Python; the discipline is language-agnostic.

    5 min
  • essayMay 10, 2025
    The middle between a prototype and a product.

    Where most AI software goes to die — the territory between a convincing demo and a system people rely on for their jobs — and what it takes to survive the crossing without throwing away everything that made the prototype fast, weird, and worth shipping in the first place.

    10 min
Read the full desk →
§ 03 — The DossierFile No. 001

For hire, with taste.

A decade-plus shipping enterprise software at Fortune 1 scale, with a through-line of systems that outlive their first ambitions. The day job is Walmart Global Tech, where I lead engineering on GenAI adoption across procurement, operations, and ERP modernization programs. Independently, SAALR is the studio I run for the same class of problem, applied to teams that do not have a Walmart to do it for them.

The projects on this page span a wider lens: an electronics PLM built for the way hardware actually ships, a graph-native wiki in pre-launch, a recruiting workflow for lean teams, an options-trading academy that commits to free forever, a learning platform for campus-scale tutoring, a peer-to-peer skills marketplace, and a pair of open-source studies in agent orchestration and voice. Some are shipping; some are pre-launch; all are serious.

IIM Calcutta alum, with certifications across AWS DevOps and Databricks, a posture I keep current because the shape of “enterprise AI” keeps moving underneath the industry it is being sold into. My bias is against software that announces itself: I’d rather ship an interface that disappears into a workflow than a dashboard that demands attention. The boring parts — auth, ergonomics, error shapes, the tiny decisions nobody will screenshot for a deck — are where the real craft shows up.

Based in San Jose. Open to advisory and select engagements starting Q3 2026 — typically AI strategy, enterprise modernization programs, and engineering organization design. For everything else, the email and GitHub below are the fastest path.