Mahendra Yadav

Senior Software Engineer

• Led a team of six on competitive-intelligence work across unstructured data sources. The tooling cut manual research time by ~30% and opened up coverage of 5 new data domains.
• Built the internal platform API around Snowflake Cortex Analyst and Agent as a shared, governed service. Roughly 8 internal teams use it, and it handles ~50,000 requests/day.
• Worked across teams to integrate the API with enterprise AI, agentic workflows, and Microsoft Teams, taking each integration through the usual contract, security, and scaling review.
• Reviewed code and raised the team's engineering bar, and broke epics into work that actually fit a sprint, which tightened predictability by ~20%.
• Ran the infrastructure roadmap for a big-data platform and restructured legacy components, saving roughly 10 hours a week in operational overhead.
• Built an alerting service from scratch. Incident detection went from ~45 minutes to under 5 minutes.
• Mentored three engineers, all of whom moved into senior individual-contributor roles.

Application Developer (Lead Consultant)

• Owned the infrastructure for a national health data platform serving ~500,000 users.
• Built Go services for access management and reporting that became standard tooling across the engineering org.
• Moved to a hub-and-spoke Azure Databricks architecture, cutting cross-workspace compute by ~20% and saving ~$8,000/month in Azure spend.
• Ran the security review on a data-catalog application and closed 12 critical findings.
• Fixed stability problems in an HDInsight Spark cluster that were causing job failures, dropping the failure rate by ~40%.
• Set up centralized logging and alerting in Azure Monitor, which cut mean time to detection (MTTD) by ~50%.
• Ran an HDInsight Kafka cluster with Debezium for real-time CDC streaming, processing ~2 million events/day.
• Hardened Kubernetes clusters using Calico network policies to shrink the pod-to-pod attack surface.
• Built batch pipelines in Python, Airflow, HDInsight, and Spark that process ~5 TB of data a day.
• Mentored new joiners on infrastructure and data engineering.

• Built CI/CD pipelines for data and ML work, cutting model deployment from ~3 days to a few hours.
• Standardized how data-science teams trained and deployed models, using CD4ML practices.
• Used MLflow, DVC, Jenkins, and HDFS for versioned storage and experiment tracking.

• Worked with the team to secure PII across the DevOps pipeline, for healthcare data-protection compliance.
• Set up AWS Data Migration Service to provision sanitized, non-PII data for consumer-facing access, which unblocked a few downstream analytics use cases.

Full Stack Developer

• Built microservices for an ad-tech campaign platform in Python, Django, gRPC, and React, supporting ~150 active campaigns a day.
• Wrote dynamic provisioning scripts on GCP so teams could create their own test environments, cutting provisioning from ~4 hours to ~15 minutes.
• Ran provisioning and operations for dev and test environments.
• Designed a Go and gRPC notification service that handles ~100,000 notifications/day for campaign and booking status alerts.

Jr Software Engineer

• Automated data ingestion and ETL with Python and Airflow, saving ~15 hours/week of manual work.
• Built Elixir/Phoenix microservices for a hospitality client, supporting ~500 concurrent users.