03 / Expertise
Systems expertise
Palash works across software architecture, cloud engineering, and applied AI. The questions change with the work: where to draw a boundary, how a service recovers, or what evidence makes an AI result useful.
Software Architecture
Palash starts with the domain and its constraints. That gives teams a basis for deciding which interfaces to expose and where to keep complexity, and for testing those decisions before revisiting them.
- System design
- API contracts
- Event-driven systems
Cloud Engineering
How a service will be deployed, monitored, and recovered shapes his design from the start. Cost belongs in that conversation too, before the system ships.
- Cloud-native design
- Infrastructure as code
- Observability
Applied AI
What needs to improve, and how would a team know it had? Those questions guide where Palash uses AI, what evidence to look for, and when a person should make the final call.
- Grounded AI
- Evaluation
- Human oversight
Capability matrix
- Programming
- Go, Python, Node.js, C#, Shell
Backend APIs in Go, data pipelines in Python, and earlier enterprise applications in C#.
- Google Cloud
- Cloud Run, Pub/Sub, BigQuery, Firestore, Dataflow, Cloud SQL
Cloud services and event-driven workflows, including pipelines for sales and interaction data.
- Cloud architecture
- Google Cloud, Alibaba Cloud
Cloud-native system design and prior work on a migration to Alibaba Cloud.
- Infrastructure
- Terraform, GitHub Actions, Cloud Build
Infrastructure as code and build and delivery automation for backend services.
- Databases
- PostgreSQL, Spanner, Oracle, SQL Server
Data persistence across backend and enterprise systems, including retail marketing data in Oracle.
- Data engineering
- Dataflow, BigQuery, data warehouses
Processing sales and interaction data in cloud pipelines and supporting retail reporting and planning.
- Enterprise systems
- ERP, CRM, retail customer data
Enterprise application development and retail data work supporting loyalty and business planning.
- AI engineering
- Review summarization, topic generation, image moderation
Palash contributed to review summarization and laid the groundwork for identifying recurring topics in customer reviews. He also built image label detection with asynchronous pre-checks to support moderation.