01
Discovery
Map the data landscape, ownership gaps, and access patterns before writing code.
Platform engineer building governed data platforms, AI agents, and access governance systems
India (IST, UTC+5:30) · Open to remote / updated September 2026
Platform Engineering.AI Agents & Governance.Production AI Systems.
12+ years building multi-agent systems, context engineering, LLM orchestration, MCP integrations, RAG pipelines, AI observability, and governed data platforms for enterprise.
Context EngineeringMulti-Agent SystemsLLM OrchestrationMCPRAGAgent HarnessAI ObservabilityEvalsTool UseProduction AI
12+ years
Experience
~5T events/day
Scale
40% faster onboarding, 30% lower cost
Measured impact
LangGraph, MCP, RAG, Prompt Engineering, FastAPI, OPA
AI stack
Selected work
Systems I have actually built.
Internal names stay private. Each writeup focuses on the problem, my ownership, the architecture, the tradeoffs, and what can be shared publicly.
Selected system
Governed conversational data platform
A multi-LLM conversational data agent that retrieves business knowledge, runs governed queries, reasons over answers, and builds charts.
LangGraphMCP (Model Context Protocol)PythonFastAPIContext Engineering
Open case studySelected system
Access governance platform
Sole architect and lead developer of an enterprise access governance platform: centralized attribute management, policy authoring for OPA/Ranger/OpenFGA, cryptographically signed bundle distribution, and enforcement that keeps working with the control plane offline.
OPA (Open Policy Agent)Apache RangerPostgresKubernetes (OCP)DataHub
Open case studySelected system
Telecom network data lake
Petabyte-scale network data lake and multi-level analytics for mobile tower events.
FlinkKafkaJavaElastic StackInflux
Open case studyHow I work
Discovery-first delivery for AI and data platforms.
The AI journey starts with discovery. I usually need to understand the data landscape, permissions, and operating constraints before architecture or model choices are worth debating.
02
Architecture
Design metadata, authorization, orchestration, and agent workflows as one governed system.
03
Development
Build with governance, observability, and tests from the first commit.
04
Delivery
Cloud-native deployment with CI/CD, monitoring, and incremental rollout.
05
Evolution
Tune models, expand coverage, close governance gaps, improve developer experience.
Tech stack
Tools I reach for when the system has to hold up in production.
Grouped the way the work actually gets done: languages, AI stack, data systems, infrastructure, databases, and cloud foundations.
Languages
PythonJavaTypeScriptScalaSQL
AI & LLM
LangGraphMCP (Model Context Protocol)RAGContext EngineeringFastAPILangfusePrompt Engineering
Data Platform
KafkaFlinkSparkTrinodbtAirflowDataHubApache HiveApache Iceberg
Infrastructure
Kubernetes (OCP)DockerOPA (Open Policy Agent)GrafanaElastic Stack
Databases
PostgresMongoDBApache HudiAlluxio
Cloud
Cloud (GCP, AWS, Azure)
Experience
Built across AI, data platforms, telecom, and retail.
I usually work on systems that already matter in production: heavy data movement, unclear ownership, fragile access paths, and platform work that has to become easier to operate.
Airtel Digital
2021 - present
Data Platform Engineer / Architect
I architect and build governed data platforms, metadata services, workflow orchestration, access-governance integrations, in-house CI/CD onboarding, and network-scale analytics systems.
dunnhumby
2018 - 2021
Data Science Engineer
I worked between data science and big-data platform teams, turning statistical and ML analysis into reusable pipelines, data marts, reporting products, and client-ready analytics workflows.
Mphasis
2014 - 2018
Software Engineer
I worked on mainframe and big-data systems for insurance and telecom clients, including Spark migration work and automation that saved 500+ manual hours per year.
Why Vijay
Built for systems that need to work after the demo.
The work stays close to production reality: platform design, implementation, and governance decisions in the same lane instead of split across separate roles.
12+ years across telecom, retail, and enterprise data, not just demos
The work spans production data platforms, internal developer systems, and AI workflows that have to survive real operational pressure.
Architecture to code: I design the system and build it
I do the system design, the service boundaries, the policy model, and the implementation details needed to make the platform actually operate.
Governance-first: authorization, metadata, and audit from day one
The platform shape starts with access, lineage, and observability so AI and data workflows stay usable after the prototype phase.