close

DEV Community

Jordan Huang profile picture

Jordan Huang

Exploring the intersection of AI and web development.

Location Sydney, Australia Joined Joined on 
Your Agent Didn't Fail — Your Server Did: Four Free-Tier Myths and a Checkpoint Contract

Your Agent Didn't Fail — Your Server Did: Four Free-Tier Myths and a Checkpoint Contract

Comments
5 min read
The Disk Isn't Yours: Four Ephemeral-Host Myths and a Restart Drill

The Disk Isn't Yours: Four Ephemeral-Host Myths and a Restart Drill

Comments
5 min read
Your Prompt Is Not Your Request: A Parity FAQ for Model Calls

Your Prompt Is Not Your Request: A Parity FAQ for Model Calls

Comments
4 min read
Burst Now, Starve Later: A Dual-Bucket Gate for Free-Tier Rate Limits

Burst Now, Starve Later: A Dual-Bucket Gate for Free-Tier Rate Limits

Comments
5 min read
Retries Don't Fix a Flaky Dependency: Five Beliefs and an Amplification Probe

Retries Don't Fix a Flaky Dependency: Five Beliefs and an Amplification Probe

Comments
5 min read
You Aborted the Request. Did the Work Stop? Four Cancellation Myths and a Retry-Budget Probe

You Aborted the Request. Did the Work Stop? Four Cancellation Myths and a Retry-Budget Probe

Comments
4 min read
Retries Are Not Free: Five Myths, an Attempt Ledger, and a Stop Rule

Retries Are Not Free: Five Myths, an Attempt Ledger, and a Stop Rule

Comments
6 min read
A Free Endpoint Isn't Slow — It's Queued: A Little's Law Load Probe

A Free Endpoint Isn't Slow — It's Queued: A Little's Law Load Probe

Comments
5 min read
Cost per Green Run: Seven Free-Tier Myths and a Ledger You Can Run

Cost per Green Run: Seven Free-Tier Myths and a Ledger You Can Run

Comments
6 min read
Retry Is Not Resilience: Five Free-Tier Model Myths and a Retry Ledger You Can Run

Retry Is Not Resilience: Five Free-Tier Model Myths and a Retry Ledger You Can Run

BERJAYA 1
Comments
5 min read
Your Process Is Up; Your Service Isn't: Five Free-Tier Uptime Myths and a Restart Probe

Your Process Is Up; Your Service Isn't: Five Free-Tier Uptime Myths and a Restart Probe

BERJAYA 1
Comments
5 min read
The First Token Is Not the Answer: Four Streaming Myths and a Chunk-Timeline Probe

The First Token Is Not the Answer: Four Streaming Myths and a Chunk-Timeline Probe

Comments
5 min read
Your Smoke Test Is Not an Eval: Five Myths and a Golden-Set Gate

Your Smoke Test Is Not an Eval: Five Myths and a Golden-Set Gate

BERJAYA 1
Comments
5 min read
Your Free Server Is Not Your Laptop: A Parity Probe for Five Assumptions That Break on Deploy

Your Free Server Is Not Your Laptop: A Parity Probe for Five Assumptions That Break on Deploy

Comments
5 min read
A 200 OK Is Not a Ready Model: Four Health-Check Myths and a One-Token Probe

A 200 OK Is Not a Ready Model: Four Health-Check Myths and a One-Token Probe

Comments
5 min read
Your Free Server Forgets Everything: Five Ephemeral-Deployment Myths and a State Probe

Your Free Server Forgets Everything: Five Ephemeral-Deployment Myths and a State Probe

Comments
5 min read
A Free Server Is Not a Small Server: Four Myths and a Boot Probe

A Free Server Is Not a Small Server: Four Myths and a Boot Probe

Comments
6 min read
Free Tier Is a Budget, Not a Promise: Four Myths and a Capability Card Probe

Free Tier Is a Budget, Not a Promise: Four Myths and a Capability Card Probe

Comments
5 min read
A Cron Entry Is Not a Job: Five Overnight-Batch Myths and a Run Ledger

A Cron Entry Is Not a Job: Five Overnight-Batch Myths and a Run Ledger

Comments
5 min read
Your Timer Is a Wall Clock, Not a Stopwatch: Four Elapsed-Time Myths and a Monotonic Probe

Your Timer Is a Wall Clock, Not a Stopwatch: Four Elapsed-Time Myths and a Monotonic Probe

Comments
5 min read
A Retry Is Not a Repeat: Five Idempotency Myths Behind a Model-Backed Endpoint

A Retry Is Not a Repeat: Five Idempotency Myths Behind a Model-Backed Endpoint

Comments 1
6 min read
Cron Fires; It Does Not Queue: Four Scheduling Myths and a Run Wrapper

Cron Fires; It Does Not Queue: Four Scheduling Myths and a Run Wrapper

Comments
5 min read
Temperature 0 Is Not Determinism: Four Sampling Myths and a Repeat-Run Probe

Temperature 0 Is Not Determinism: Four Sampling Myths and a Repeat-Run Probe

Comments 1
5 min read
Same Model Name, Different Answers: A Drift Probe for Free Model Endpoints

Same Model Name, Different Answers: A Drift Probe for Free Model Endpoints

Comments 1
6 min read
Free Tokens Aren't Free Requests: Five Budget Myths and a Spend Ledger You Can Run

Free Tokens Aren't Free Requests: Five Budget Myths and a Spend Ledger You Can Run

Comments
4 min read
Your JSON Mode Is Not a Schema: Four Structured-Output Myths and a Validation Gate

Your JSON Mode Is Not a Schema: Four Structured-Output Myths and a Validation Gate

BERJAYA 1
Comments 1
5 min read
A Fallback Model Is a Stranger: Four FAQ Answers Before You Route Traffic to It

A Fallback Model Is a Stranger: Four FAQ Answers Before You Route Traffic to It

Comments
5 min read
Your Free Server's Local Disk Is Not a Database: Five Persistence Myths and a Restart Probe

Your Free Server's Local Disk Is Not a Database: Five Persistence Myths and a Restart Probe

Comments
5 min read
A Restart Is Not a Reset: Five Persistence Myths and a Boot-ID Probe

A Restart Is Not a Reset: Five Persistence Myths and a Boot-ID Probe

Comments
4 min read
Your Free Tier Is Not a Capacity Plan: Five Throughput Myths and a Concurrency Ramp

Your Free Tier Is Not a Capacity Plan: Five Throughput Myths and a Concurrency Ramp

Comments
5 min read
Streaming Feels Faster, but Your Model Isn't: A Wire-Level Timestamp FAQ

Streaming Feels Faster, but Your Model Isn't: A Wire-Level Timestamp FAQ

BERJAYA 1
Comments 1
5 min read
Valid JSON, Wrong Order: Five Structured-Output Myths and a Three-Gate Replay Test

Valid JSON, Wrong Order: Five Structured-Output Myths and a Three-Gate Replay Test

Comments
5 min read
Your Cache Key Is Not Your Prompt: Four Response-Cache Myths and a Collision Test

Your Cache Key Is Not Your Prompt: Four Response-Cache Myths and a Collision Test

Comments 1
5 min read
Four Idle Myths About Free Servers, and the Wake Probe That Settles Them

Four Idle Myths About Free Servers, and the Wake Probe That Settles Them

Comments
6 min read
Your Timeout Is Local: Five Myths About Model Calls You Retry

Your Timeout Is Local: Five Myths About Model Calls You Retry

Comments
7 min read
Your App Is Up, Your Model Call Is Down: Five Coupling Myths and a Two-Probe Check

Your App Is Up, Your Model Call Is Down: Five Coupling Myths and a Two-Probe Check

Comments
6 min read
A 200 OK Is Not a Finished Answer: Classifying Model-Call Failures Before You Retry

A 200 OK Is Not a Finished Answer: Classifying Model-Call Failures Before You Retry

Comments
6 min read
Same Prompt, Different Answer: Five Reproducibility Myths in Free-Model Pipelines

Same Prompt, Different Answer: Five Reproducibility Myths in Free-Model Pipelines

Comments
5 min read
Your Free Server Forgot Your Files: Four Restart Myths and a Survival Check

Your Free Server Forgot Your Files: Four Restart Myths and a Survival Check

Comments
5 min read
Same URL, Different Machine: Four State Myths on a Free Server

Same URL, Different Machine: Four State Myths on a Free Server

Comments
5 min read
nohup Is Not a Checkpoint: An FAQ on Batches That Outlive Your SSH Session

nohup Is Not a Checkpoint: An FAQ on Batches That Outlive Your SSH Session

Comments
5 min read
Your Health Check Was Green. The Server Was Asleep.

Your Health Check Was Green. The Server Was Asleep.

Comments
6 min read
The Free Server Slept Overnight: Four State Claims Your App Can't Back Up

The Free Server Slept Overnight: Four State Claims Your App Can't Back Up

Comments
5 min read
A Bigger Context Window Is Not More Memory: Five Token Myths and a Budget Gate

A Bigger Context Window Is Not More Memory: Five Token Myths and a Budget Gate

BERJAYA 1
Comments 2
5 min read
200 OK Proves Nothing About Tool Calls: A Free-Model Capability FAQ With a Probe

200 OK Proves Nothing About Tool Calls: A Free-Model Capability FAQ With a Probe

Comments
5 min read
It Passed CI. That Still Doesn't Make the Patch Safe.

It Passed CI. That Still Doesn't Make the Patch Safe.

Comments
6 min read
A Free Model Swap Is Not a Config Change: A Contract Probe You Can Run

A Free Model Swap Is Not a Config Change: A Contract Probe You Can Run

Comments
6 min read
Free Tokens Are Not Free Work: Five Quota Myths and a Ledger You Can Run

Free Tokens Are Not Free Work: Five Quota Myths and a Ledger You Can Run

Comments
5 min read
A 429 Is Not Always a Rate Limit: Four Free-Tier Failure Myths and a Classifier You Can Run

A 429 Is Not Always a Rate Limit: Four Free-Tier Failure Myths and a Classifier You Can Run

Comments
6 min read
UTC Is Not Agreement: Five Clock Myths That Break Deploys, Plus a Drift Probe

UTC Is Not Agreement: Five Clock Myths That Break Deploys, Plus a Drift Probe

Comments
6 min read
Temperature 0 Is Not a Reproducibility Setting: Five Model Myths and a Drift Probe

Temperature 0 Is Not a Reproducibility Setting: Five Model Myths and a Drift Probe

Comments
5 min read
Your Code Ran on the Free Box. That's Not an Environment Match.

Your Code Ran on the Free Box. That's Not an Environment Match.

Comments
5 min read
The Free Box Is Not Your Laptop: Five Portability Myths and a Probe You Can Diff

The Free Box Is Not Your Laptop: Five Portability Myths and a Probe You Can Diff

Comments
6 min read
Your Laptop Is Not the Server: An Environment Parity FAQ for Free-Tier Boxes

Your Laptop Is Not the Server: An Environment Parity FAQ for Free-Tier Boxes

Comments
5 min read
Your Free Server Forgets Everything: Six Deploy Myths and a Restart Drill

Your Free Server Forgets Everything: Six Deploy Myths and a Restart Drill

Comments
5 min read
Free Server, Free Model: Four Assumptions That Break Your First Deploy (And a Probe Script)

Free Server, Free Model: Four Assumptions That Break Your First Deploy (And a Probe Script)

Comments
5 min read
The Context Window Is a Budget, Not a Memory: An FAQ With a Truncation Ledger

The Context Window Is a Budget, Not a Memory: An FAQ With a Truncation Ledger

Comments
5 min read
Assume Nothing Persists: Five Free-Tier Myths and a Restart Drill You Can Re-Run

Assume Nothing Persists: Five Free-Tier Myths and a Restart Drill You Can Re-Run

Comments
5 min read
A Timeout Is Not a Slow Model: Five Latency Myths and a 40-Line Attribution Script

A Timeout Is Not a Slow Model: Five Latency Myths and a 40-Line Attribution Script

Comments
4 min read
Pin the Model, Log the Tokens: An FAQ for Free-Tier Agent Runs

Pin the Model, Log the Tokens: An FAQ for Free-Tier Agent Runs

Comments
6 min read
loading...