SSSamarth Sinha
Résumé

Data & AI Solutions Architect · 14 years

Samarth Sinha I design data platforms and AI systems that hold up in production.

Senior Solution Architect at Searce, a Google Cloud partner. Before that I led the data-platform function at Tokopedia and ran a 4.2 PB Hadoop platform at Airtel.

Explore my work Request résumé
data-platform.architectureIllustrative
How the pieces of a modern data platform fit together Applications to Ingestion. Files and SaaS to Ingestion. Legacy estates to Ingestion. Ingestion to Lakehouse. Governance to Lakehouse. FinOps to Lakehouse. Lakehouse to AI agents. Lakehouse to Analytics. Lakehouse to Data products. Applications events and CDC Files and SaaS batch feeds Legacy estates Oracle, Hadoop Ingestion streaming + batch Governance contracts, quality Lakehouse BigQuery · Delta FinOps cost per team AI agents human approval Analytics trusted metrics Data products owned by domains
Sources→Governed lakehouse→AI & analytics
14 yrsin data engineering and architecture
4.2 PBHadoop platform set up and run
34domain-owned data projects, up from 4
32–43%lower cloud spend, per team
11direct reports across five teams

Where I've built and run data platforms

  • Searce
  • Tech Mahindra
  • Tokopedia
  • Bharti Airtel
  • ITC Infotech
  • Elitecore
  • CRMnext

About

Fourteen years, from database rooms to AI systems

I'm a data and AI solutions architect with fourteen years in data. I started as an Oracle DBA keeping high-availability telecom databases running, then set up and ran a 4.2 PB Hadoop platform at Airtel.

At Tokopedia (GoTo Group) I led the data-platform function: eleven direct reports across five teams. We moved a monolithic platform to a domain data mesh of 34 domain-owned projects, made data quality visible to every team, and cut cloud spend 32–43% per team.

Today I'm a Senior Solution Architect at Searce, a Google Cloud partner, where I design data-platform modernisation and AI programmes for enterprises: discovery, migration strategy, lakehouse architecture and disaster recovery.

I also build AI systems hands-on to learn what actually holds up, such as agents behind human approval gates, blind evaluation harnesses and reproducible data platforms, and I write up what I learn as architecture notes.

Capabilities

What I bring to a data platform

Eight areas, each backed by work I've actually done.

Data architecture & data mesh

Lakehouse and domain-oriented platforms with governance built in.

  • Monolith → 34 domain-owned GCP projects
  • Shared policies for replication, quality, catalog, access
  • 17-parameter source-quality rating with team alerts

Cloud modernisation

Moving estates to Google Cloud without stopping the business.

  • Migration strategy: waves, parallel runs, cutover design
  • Lakehouse design on BigQuery and Databricks
  • Cross-cloud DR with Delta Sharing, Azure → Google Cloud

Pipelines & streaming

Batch and streaming pipelines that teams can actually operate.

  • Kafka, Spark and Airflow platforms at Tokopedia
  • Azure Data Factory + Databricks pipelines
  • Ab Initio and NiFi ETL into a 4.2 PB Hadoop platform

Governance & data quality

Rules that are enforced, not just written down.

  • Data-access matrix with scheduled audits
  • PII and GDPR controls
  • Data contracts and schema-drift handling

FinOps

Cost treated as an architecture decision, not a monthly surprise.

  • 32–43% lower cloud spend per team
  • Cost attribution to business units
  • 25+ cost and service-level reports

AI & agentic systems

AI that is governed, evaluated and cost-aware.

  • Agents behind contracts, risk tiers and human approval
  • Blind, repeated LLM evaluation before switching models
  • AI-readiness assessment of data estates

Engineering leadership

Teams that own their platforms, end to end.

  • 11 direct reports across five teams
  • OKRs and a nine-step performance-review process
  • Engineers developed into leads

Solution consulting

From the first workshop to a statement of work.

  • Discovery, sizing and consumption forecasts
  • Effort estimates and SOWs
  • Multi-round architecture reviews with customers

How I work

  1. 01

    Discover

    Workshops and estate sizing: what the data has to do for the business, what exists today and what it costs.

  2. 02

    Design

    Target architecture, migration waves, governance and recovery, with the trade-offs written down.

  3. 03

    Prove

    PoCs, parity checks and evaluations before anyone commits, so a design gets tested, not just believed.

  4. 04

    Operate

    Run what ships: cost attribution, quality ratings, access audits and clear ownership.

Selected work

Platforms I've led, built and run

Short engineering reports: the context, what I did, the architecture and what changed. Each one says how far the work got.

Tokopedia (GoTo Group) · 2019 — 2024

From a Monolithic Data Platform to a Domain Data Mesh

Led the move of a large e-commerce company's central data platform into 34 domain-owned cloud projects, with shared policies and a source-quality rating.

4 → 34domain-owned GCP projects
11direct reports across five teams
17parameter source-quality rating
  • Google Cloud
  • BigQuery
  • GKE
  • Airflow
  • Kafka
Read the case study

Tokopedia (GoTo Group) · 2019 — 2024

Cloud FinOps Programme Across Five Teams

Ran a four-phase FinOps programme across five teams that cut cloud spend 32–43% per team, sustained by continuous cost monitoring.

32–43%lower cloud spend per team
5teams in the programme
25+cost and service-level reports
  • Google Cloud
  • BigQuery
  • Cost attribution
  • Data Studio
  • Grafana
Read the case study

Bharti Airtel · 2018 — 2019

Running a 4.2 PB Hadoop Platform

Set up and ran a 4.2 PB Cloudera Hadoop platform end to end, with ETL, reporting and automation, plus two Cassandra clusters for Customer 360.

4.2 PBCloudera Hadoop platform
15+cluster-utilisation reports
2Cassandra clusters for Customer 360
  • Cloudera Hadoop
  • Hive
  • YARN
  • Ab Initio
  • NiFi
Read the case study

All case studies

Experience

Career journey

Seven roles across telecom, HR services, e-commerce and cloud consulting.

  1. CRMnext
  2. Elitecore
  3. ITC Infotech
  4. Airtel
  5. Tokopedia
  6. TechM
  7. Searce
  1. Databases
  2. Big data
  3. Cloud data platforms
  4. Data & AI
20122014201620182020202220242026

  1. 2025 — Present

    Senior Solution Architect · Searce

    Architecture and pre-sales for enterprise data-platform modernisation and AI programmes on Google Cloud: discovery, sizing, migration strategy, lakehouse design and cross-cloud disaster recovery.

    • BigQuery
    • Databricks
    • Delta Lake
    • Kafka
    • Azure
    • Terraform

  2. 2024 — 2025

    Tech Specialist · Tech Mahindra

    Databricks architect on Azure. Designed complete, end-to-end data pipelines with Azure Data Factory and Databricks.

    • Azure Data Factory
    • Databricks
    • Azure

  3. 2019 — 2024

    Data Engineer Senior Lead · Tokopedia (GoTo Group)

    Lead Data Platform → Engineering Manager, Data Platform

    Led the data-platform function: 11 direct reports across data engineering, data science, data analytics and two internal teams. Moved a monolithic platform to a domain data mesh (4 → 34 domain-owned GCP projects), ran a four-phase FinOps programme that cut cloud spend 32–43% per team, and built a 17-parameter data-quality rating with alerts by team and service.

    • Google Cloud
    • BigQuery
    • GKE
    • Airflow
    • Kafka
    • Spark
    • Terraform

  4. 2018 — 2019

    Sr Data Admin · Bharti Airtel

    Set up and ran a 4.2 PB Cloudera Hadoop platform end to end, with Ab Initio and NiFi ETL, billing and utilisation reporting, and Ansible and Grafana automation. Stood up two Cassandra clusters for Customer 360.

    • Hadoop
    • Hive
    • YARN
    • NiFi
    • Cassandra
    • Ansible

  5. 2016 — 2018

    Data Analyst · ITC Infotech

    Database engineering and a Hadoop analytics warehouse for a global HR-services client; migrated data from SAP IQ to Hadoop.

    • Hadoop
    • Hive
    • Oozie
    • SAP IQ

  6. 2014 — 2016

    Oracle DBA · Elitecore Technologies

    L3 DBA for high-availability telecom platforms: 12 production RAC databases and 32 migrations for one operator, plus Data Guard and database builds for operators in Dubai and Malaysia.

    • Oracle RAC
    • Data Guard
    • GoldenGate

  7. 2012 — 2014

    Oracle DBA · CRMnext

    Oracle and SQL Server administration and PL/SQL development for a banking CRM platform.

    • Oracle
    • SQL Server
    • PL/SQL

Technology timeline

When I used what, taken from the roles above.

Oracle & SQL Server
Hadoop · Hive · NiFi
Cassandra
Google Cloud · BigQuery · GKE
Kafka · Spark · Airflow
Terraform · CI/CD
Azure · Data Factory · Databricks
LLMs & agentic systems
20122014201620182020202220242026

Notes

Architecture notes

Architecture notes: patterns I design with, and lessons from building data and AI systems.

All notes

Toolkit

What I build with

Cloud

  • Google Cloud
  • BigQuery
  • Dataflow
  • Dataproc
  • Dataplex
  • Pub/Sub
  • GKE
  • Cloud SQL
  • Secret Manager
  • Azure
  • Azure Data Factory
  • Databricks

Data

  • Spark
  • Kafka
  • Airflow
  • Delta Lake
  • Delta Sharing
  • Unity Catalog
  • Hadoop
  • Hive
  • NiFi
  • Cassandra
  • Oracle RAC
  • PostgreSQL
  • SQL Server
  • DuckDB

Architecture

  • Data mesh
  • Lakehouse
  • Data contracts
  • Data governance
  • Data quality
  • Change data capture
  • FinOps
  • Disaster recovery
  • Migration strategy

AI

  • Agentic systems
  • MCP
  • LLM evaluation
  • Model routing
  • Human-in-the-loop
  • Ollama
  • Gemini
  • Claude
  • GPT

Platform

  • Terraform
  • Kubernetes
  • GitHub Actions
  • Jenkins
  • Ansible
  • Grafana
  • Looker Studio
  • Tableau

Languages

  • Python
  • SQL
  • PL/SQL
  • Shell
  • B.Tech, Computer Science & Engineering · Maharshi Dayanand University, 2012
  • Oracle Certified Associate · Oracle Database 11g
  • Winner, "India's Top 10 Ethical Hackers" contest · IIT Kanpur, 2011

Contact

Have a data platform or an AI system to get right?

I'm happy to talk architecture, modernisation or governed AI. Email is quickest.

samarthsinhajob@gmail.com LinkedIn Request résumé

↑↓ to move · Enter to open · Esc to close