DynoTable
Blog

DynoTable Engineering Blog

How DynoTable is built — the engineering stories behind the AI agent, the MCP server, and the rest of the product.

These are working notes from building DynoTable, a desktop DynamoDB client. Each post walks through one engineering problem we shipped through for real — staging every write behind a reviewable diff, running an OAuth server inside an Electron app, making a local SQL engine respect DynamoDB’s access-pattern rules.

We write them from the commits and the design notes — expect concrete constraints, the decisions we reversed, and code where it matters.

If you want DynamoDB fundamentals rather than build stories, the Learn section is the structured course — indexes, single-table design, capacity math and the rest.

· 8 min read

How many writes per second before a new DynamoDB table throttles?
We drove a fresh on-demand table until it throttled — 4,130 writes/s, flat, run after run, reads past 12,700/s untouched. Method, raw numbers, growth curve.

· 7 min read

Why we hand-wrote a PartiQL parser for DynamoDB
Why we hand-wrote a PartiQL parser for DynamoDB instead of a grammar generator — error-tolerant parsing, a 208-fixture AWS spec corpus. MIT, open source.

· 9 min read

Your agent doesn't need every tool in its context
Our AI agent can reach 38 tools and almost never sees them all: an always-visible core, a searchable tail, and the line cheap models forced us to draw.

· 9 min read

GPT-5.6 Luna vs other cheap models: benchmarked on real agent work
We benchmarked GPT-5.6 Luna against 11 models — Claude Haiku 4.5, Nova 2 Lite, DeepSeek V3.2 and more — on real agent evals, with per-call Bedrock costs.

· 8 min read

How we built a local SQL engine for DynamoDB
DynamoDB has no JOIN, no GROUP BY, no COUNT(*). How we built a local SQL engine that runs real SELECT queries with joins and aggregates over your tables.

· 8 min read

Why a database GUI should never write directly
Why DynoTable doesn't write to DynamoDB on Save. An item edit becomes a reviewable per-attribute diff and ships as a transaction guarded by optimistic locks.

· 9 min read

What it took to build a safe DynamoDB MCP server
How we built a safe DynamoDB MCP server. An OAuth 2.1 authorization server runs inside the Electron app, and each connection binds to one AWS profile.

· 10 min read

How we made cheap LLMs reliable at DynamoDB
How we made cheap LLMs reliable at DynamoDB. Rewriting one tool validator's error message took Nova Lite from 0% to 100% on a join eval, no fine-tuning.

More DynamoDB resources

Learn DynamoDB
A structured DynamoDB curriculum, from the fundamentals through to the internals.
Common DynamoDB Errors
Every common DynamoDB error with its cause and the exact fix.
DynamoDB Code Examples
Copy-paste snippets for every operation — SDK v3, boto3, and the AWS CLI.
DynamoDB FAQ
Straight answers to the DynamoDB questions developers ask most.
DynamoDB Tools
Expression builder, pricing and capacity calculators — right in your browser.

Work with DynamoDB without the Console

A fast DynamoDB desktop client that runs the real SQL DynamoDB can’t — JOINs, GROUP BY, aggregates — with visual editing and an AI agent on your own Bedrock keys.

DownloadSee pricing

Free 30-day trial, no credit card — then the Free plan with no time limit.