Work

What Prasad has architected and built.

Selected systems, prototypes, research-backed architectures and public evidence. Repository inventory remains available under GitHub Engineering.

5 selected work items shown

Governed SAP engineeringSAP, ABAP Cloud, agents, evidence gates

CAPELLA AFEP

Purpose: SAP engineering needs evidence, authority boundaries and controlled execution instead of prompt-only automation.
Outcome: Agentic Full-Stack Engineering Platform for SAP, with public evidence for governed delivery.

Decision intelligenceAstro, TypeScript, deterministic graph runtime

Purpose: Architecture judgment is hard to inspect, compare or challenge from outside the human decision process.
Outcome: Evidence, constraints, counterfactuals, fidelity and receipts are exposed as inspectable decision surfaces.

Architecture IntelligenceAstro, JSON contracts, Playwright QA

Purpose: Enterprise architecture decisions need traceable scenarios, evidence and learning surfaces.
Outcome: Scenario Studio, Evidence Explorer, architecture graph, Learn and public-safe demonstrations.

SAP delivery control planeABAP Cloud, RAP, Eclipse ADT, SAP constraints

Purpose: ABAP Cloud delivery teams need a standards-led automation path that can move from an AI or coding agent through Eclipse ADT MCP tooling into SAP ADT execution while preserving audit evidence and clear proof boundaries.
Outcome: Validated architecture option for Clean Core ABAP on Cloud delivery.

Enterprise AI controlAgents, SAP service mapping, telemetry, audit

Purpose: Enterprise teams need a governed way to configure, execute and observe AI agents around SAP work.
Outcome: Governed enterprise AI architecture for SAP-adjacent agent execution.

Complete catalogues

Work is outcome-led here. Engineering remains the repository-evidence surface.