Mahendra Yadav
Resume Email

Built to hold under load.

I build the platform APIs and event streams enterprises run on.

Email me See the work

Mahendra Yadav, Senior Software Engineer at ServiceNow, Hyderabad.
9+ years in Go and Python.

Now

A short record of where I am and what I am working on.

Portrait of Mahendra Yadav
Mahendra Yadav, Hyderabad, 2026.
Currently
Senior Software Engineer at ServiceNow, building governed platform APIs and agentic workflows on Snowflake Cortex (8 internal engineering teams, ~50,000 requests/day).
Focus
Distributed backend platforms, event-driven architecture, API design and governance, and evaluation governance for enterprise LLM agents.
Based in
Hyderabad, India.
Open to
Senior engineering roles, and distributed-systems conversations.

Last updated October 2026.

Selected systems

Five platforms I have designed and run in production. One is opened here; the other four follow in full.

National Health Data Platform

I owned the backend and infrastructure architecture for Singapore’s national health-data platform, which supported ~500,000 active users. I designed Go services for access management, identity, and reporting that became standard enterprise tooling, and ran event-driven ingestion and batch pipelines moving ~5 TB of data a day. The biggest structural change was moving to a hub-and-spoke Azure Databricks compute topology.

Impact: cut cross-workspace compute spend by ~20%, reduced distributed job failure rates by ~40%, and found ~$96K in recurring annual savings across Azure infrastructure.

Go, Python, Azure, Azure Databricks, Apache Spark, Kafka, Debezium, Terraform

500,000

active users on the national health data platform.

Thoughtworks, Health Promotion Board, Singapore · 2020 to 2024

Enterprise AI Platform API & Agentic Systems

ServiceNow · since 2024

I architected a unified, governed backend platform API around Snowflake Cortex Analyst and Cortex Agent, and run it for 8 internal engineering teams with strict latency, authentication, and throughput controls. I also integrated enterprise agentic workflows and Microsoft Teams, with secure payload validation, role-based access, and rate limiting.

Impact: handles ~50,000 requests/day and expanded automated domain coverage across 5 new business areas.

Go, Python, Snowflake Cortex, Agentic Workflows, Evals, REST, Microsoft Teams

Real-Time CDC Event Streaming (Kafka + Debezium)

Thoughtworks · data platform

I built and ran an event-driven Change Data Capture pipeline on an HDInsight Kafka cluster with Debezium connectors. It picks up transactional changes from the source databases as they happen and pushes them to downstream analytics engines and data catalogs, without adding lock contention on the primary OLTP tables.

Impact: handled ~2,000,000 CDC events/day with sub-second event propagation, supporting both continuous streaming and ~5 TB/day batch Spark pipelines.

Apache Kafka, Debezium, Event-Driven Architecture, Apache Airflow, HDInsight, Python

Kubernetes Calico Micro-Segmentation & CD4ML Pipeline

Thoughtworks · healthcare and Daimler

I locked down production Kubernetes clusters with Calico network policies, isolating namespace traffic and shrinking the pod-to-pod attack surface. On the Daimler DataOps work, I built a standard Continuous Delivery for Machine Learning pipeline with MLflow, DVC, and Jenkins so model training and deployments were reproducible and versioned.

Impact: cut the ML model deployment cycle from ~3 days to under 3 hours and resolved 12 critical security findings in the data-catalog applications.

Kubernetes, Calico, CD4ML, MLflow, DVC, Jenkins, HDFS

High-Velocity Notification Engine & GCP Automation

BeautifulCode · Sojern, travel ad-tech

I designed a scalable notification engine in Go and gRPC that streamed high-throughput campaign and booking updates across the travel ad-tech services. I also automated test-environment provisioning on Google Cloud Platform with infrastructure as code.

Impact: cut staging setup from ~4 hours to ~15 minutes and kept campaign and booking updates flowing at high throughput.

Go, gRPC, Python, Django, React, Google Cloud, Docker

How I work

Four principles I have settled on after nine years of building and running distributed systems.

I design for failure

Things break. Brokers drop, jobs stall, nodes run out of memory. I build systems that degrade gracefully and make their failures visible, instead of quietly hoping they won’t.

I treat governance as the product

When many teams share one platform, contracts, access control, and observability aren’t overhead. They are the product. I would rather build that in from the start than bolt it on later.

I price cost into the design

Efficiency is a design constraint, not a quarterly cleanup. Cost, latency, and reliability are decisions you make up front, because that is when they are still cheap to change.

I keep novelty off the critical path

I reach for proven tools and keep novelty off the critical path. The interesting complexity belongs in the problem I am solving, not in the stack that has to hold up at 3 a.m.

Roles

Nine years across healthcare, ad-tech, and enterprise software.

ServiceNow

Senior Software Engineer

Oct 2024 to present

Hyderabad, India

  • Architected and delivered a unified, governed backend platform API around Snowflake Cortex Analyst and Cortex Agent, handling ~50,000 requests/day across 8 internal engineering teams.
  • Led architectural alignment across those 8 teams to standardize API contracts, security protocols, and integration interfaces.
  • Led a six-engineer squad delivering an unstructured-data intelligence platform, with cross-team system design and automated domain coverage across 5 new business areas.
  • Established technical governance and evaluation standards with an automated Evals framework, benchmarking LLM answer quality and regression-testing model outputs.
  • Integrated enterprise agentic workflows and Microsoft Teams with the platform APIs, enforcing secure payload validation, role-based access, and rate limiting.
  • Built an end-to-end distributed alerting service from scratch. Incident detection went from ~45 minutes to under 5 minutes.
  • Drove architectural roadmapping for big-data platforms, decomposed epics into parallel work, and raised code quality through design and PR reviews.
  • Mentored three engineers in system design, distributed debugging, and production readiness.

Go, Python, Snowflake Cortex, Agentic Workflows, Evals, REST, Microsoft Teams

Thoughtworks

Application Developer (Lead Consultant)

Mar 2020 to Oct 2024

Hyderabad, India

Project: Health Promotion Board, Singapore

  • Owned the backend and infrastructure architecture for a national health-data platform supporting ~500,000 active users.
  • Designed and developed Go platform services for access management, identity, and reporting, later adopted as standard enterprise tooling.
  • Engineered event-driven data ingestion on Azure HDInsight Kafka with Debezium for real-time change data capture.
  • Built distributed batch pipelines with Python, Airflow, Kafka, and Azure Databricks (Spark), processing ~5 TB of data daily.
  • Architected a hub-and-spoke Azure Databricks compute topology, cutting infrastructure compute spend by ~20%.
  • Led root-cause analysis and performance tuning on HDInsight Spark clusters, reducing job failure rates by ~40% and stabilizing critical SLAs.
  • Hardened security for data-catalog applications, resolving 12 critical vulnerabilities and enforcing Calico Kubernetes network policies.
  • Implemented cost-governance measures across Azure infrastructure, finding ~$96K in recurring annual savings.
  • Mentored junior and mid-level consultants on distributed systems patterns, Go concurrency, and production reliability.

Project: Daimler · Data Ops & CD4ML

  • Engineered enterprise CI/CD automation and CD4ML practices, cutting model release cycles from ~3 days to under 3 hours.
  • Built reproducible data-engineering workflows with MLflow, DVC, Jenkins, and HDFS for end-to-end data auditability and experiment tracking.

Project: Telemedicine Platform

  • Architected HIPAA-compliant PII sanitization workflows with AWS DMS, provisioning secure, anonymized data pipelines for downstream analytics.

Go, Python, Azure, AWS, Kubernetes, Calico, Databricks, Kafka, Debezium, Terraform

BeautifulCode

Full Stack Developer

Mar 2018 to Mar 2020

Hyderabad, India

Project: Campaign Management Platform · Sojern (travel ad-tech)

  • Developed high-performance backend microservices with Python, Django, gRPC, and React for a travel advertising campaign platform.
  • Designed a scalable notification engine in Go and gRPC, streaming high-throughput campaign and booking updates.
  • Automated GCP test-environment provisioning with infrastructure as code, cutting staging setup from ~4 hours to ~15 minutes.

Go, Python, Django, gRPC, Google Cloud, React, Docker

Azri Solutions

Jr Software Engineer

Dec 2016 to Mar 2018

Hyderabad, India

Project: GRNConnect

  • Automated ETL ingestion pipelines with Python and Airflow, eliminating ~15 hours/week of repetitive manual operations.
  • Developed backend services with Elixir and Phoenix for high-concurrency booking endpoints.

Python, Apache Airflow, Elixir, Phoenix, ETL

Writing & speaking

Notes on distributed systems, event streaming, and cloud architecture, plus the talks I give about them.

Speaking & community

Managing Data Pipelines using Airflow

PyConf Hyderabad 2017

A talk on writing Airflow DAGs that survive failure: task retries, dependency design, and how to stop one bad task from taking down a pipeline.

Internal Tech Forums & Community Sessions

Thoughtworks & local meetups

Ran internal knowledge-sharing forums that grew to ~30 engineers, and gave talks on Docker, containerization, and database sharding.

Mentorship & Education

Dr B C Roy Engineering College, 2013 to 2017

Mentored 5 new joiners and 3 senior engineers, establishing onboarding curricula and accelerating the path to independent contribution. B.Tech in Computer Science & Engineering.

Toolkit

Tools and platforms from nine years of building and running production systems.

Technical skills

Languages

Go (Golang), Python, SQL, Node.js, Elixir

Distributed systems & architecture

Microservices, Event-Driven Architecture, High-Throughput Systems, gRPC, REST APIs, API Design & Governance, Protocol Buffers, System Design, Concurrency Models

Messaging & streaming

Apache Kafka, Debezium (CDC), Event Streaming

Data & databases

PostgreSQL, MySQL, Redis, Snowflake, Databricks (Spark), HDFS

Cloud & platforms

AWS, Azure, GCP, Kubernetes, Docker, Helm, Calico Network Policies

Enterprise AI & platform APIs

Snowflake Cortex Analyst & Agent APIs, Agentic Workflows, LangChain, LangGraph, LLM Evaluation Frameworks (Evals), Prompt Engineering

DevOps & observability

Terraform, CI/CD (Jenkins, GitLab CI, Bamboo, Argo CD), OpenTelemetry, Grafana, Jaeger, Azure Monitor, Splunk, Dynatrace

Engineering leadership & practices

Cross-Team Technical Governance, Architecture Roadmapping, EPIC Decomposition, Code Reviews, Mentorship, TDD

Certifications

Google Cloud

Professional Cloud Architect

Verify credential

Google Cloud

Associate Cloud Engineer

Verify credential

Microsoft

Azure Administrator Associate

Verify credential

Microsoft

Azure Developer Associate

Verify credential

Get in touch

Happy to talk about distributed systems, event streaming, cloud platform work, or senior engineering roles.

Email
Resume GitHub LinkedIn WordPress

Hyderabad, India · IST (UTC+5:30)

Email copied to clipboard