Notes from production
Blog
Practical articles on AWS Bedrock cost optimization, knowledge graphs, AI infrastructure, and cloud architecture — from the ThinkXL team.
AWS Bedrock vs SageMaker: How to Pick the Right One
- Pratik Kulkarni
- AWS , Cloud Architecture
- 07 May, 2026
If you're building an AI product on AWS, you'll hit this question early: Bedrock or SageMaker? The short answer is that they solve different problems, and most startups only need one. What Each Se
Read more
AWS Bedrock Cost Structure: What You're Actually Paying For
- Pratik Kulkarni
- FinOps , AWS
- 05 May, 2026
AWS Bedrock looks simple from the outside — call an API, get a response, pay per token. The reality is that a production Bedrock setup has several distinct cost layers, and they behave very differentl
Read more
Why Your AWS Bedrock Bill Makes No Sense (And How to Fix It)
- Pratik Kulkarni
- FinOps , AWS
- 30 Mar, 2026
When a startup says "our AWS bill is too high," the conversation almost always starts at the aggregate level — total monthly spend, a few large services, maybe a spike someone noticed. That's not wher
Read moreCategories
Tags
- Agentcore
- Unit economics
- Bedrock
- Cost optimization
- Finops
- Bedrock agents
- Fargate
- Ai agents
- Langgraph
- Claude code
- Mcp
- Aws agent toolkit
- Iam
- Developer tools
- Aws cur
- Sagemaker
- Ai infrastructure
- Production readiness
- Well architected
- Operational readiness review
- Amazon bedrock
- Prompt caching
- Cost optimisation
- Custom model import
- Self hosted models
- Gpu
- Aws alternatives
- Hetzner
- Ovhcloud
- Contabo
- Egress
- Neo4j
- Knowledge graph
- Terraform
- Claude
- Llm
- Hetzner vs aws
- Graviton
- Knowledge graphs
- Graphrag
- Rag
- Genai
- Aws migration
- Observability
- Postgres
- Disaster recovery
- Inference endpoints
- Evals
- Llm as judge
- Bedrock knowledge bases
- Amazon s3
- Managed knowledge base
- Agentic retrieval
- S3 vectors
- Opensearch serverless
- Vector database