Riley Li

Riley Li

Sep 21

Inventory What You Cannot Pin Before You Choose Free Agent Compute

#ai #agents #devops #python

8 min read

Riley Li

Sep 20

Budget the Agent Loop Before You Choose Shared or Isolated Compute

#ai #testing #agents #devops

7 min read

Riley Li

Sep 19

Score the Host Before You Promote a Vibe-Coded Agent

#ai #testing #python #devops

6 min read

Riley Li

Sep 18

Choose Agent Compute by Replayable Evals, Not by Spare Capacity

#ai #python #testing #devops

8 min read

Riley Li

Sep 17

Pass Four Fit Gates Before You Choose Free or Isolated Agent Compute

#ai #devops #testing #productivity

7 min read

Riley Li

Sep 16

Who Owns the Failed Agent Run? Score Free Shared Compute Against a Box You Control

#agents #ai #devops #testing

7 min read

Riley Li

Sep 14

Free Compute or Self-Hosted? A Break-Even Guide for Agent Workloads

#ai #agents #opensource #python

5 min read

Riley Li

Sep 13

Rank Agent Runtimes by the Contract You Can Enforce

#ai #agents #testing #architecture

8 min read

Riley Li

Sep 12

Invert Non-Negotiable Constraints Before You Accept Free Agent Compute

#ai #agents #architecture #python

7 min read

Riley Li

Sep 11

Keep the Agent Loop Portable, Then Choose Free Shared Compute

#ai #agents #python #architecture

7 min read

Riley Li

Sep 9

Pick Agent Compute by Blast Radius, Not by the Price Tag

#ai #agents #python #architecture

1

7 min read

Riley Li

Sep 6

Choose Agent Compute by What You Can Falsify

#ai #agents #architecture #python

7 min read

Riley Li

Sep 5

Score the Agent Before You Pick the Endpoint

#python #ai #architecture #agents

10 min read

Riley Li

Sep 4

Score the Lane First: Free Shared Workspace, Paid API, or Self-Hosted

#ai #agents #architecture #productivity

7 min read

Riley Li

Sep 4

Designing for a 10M-Token Ceiling: A Token-Budget Scheduler for Background LLM Jobs

#llm #python #architecture #ai

5 min read

Riley Li

Sep 4

A Token Bucket for Free-Tier LLM Endpoints: Rate Limiting Without a Fancy Gateway

#python #llm #api #tutorial

4 min read

Riley Li

Sep 3

Keep the Regex Writer Until Shadow Receipts Match

#python #testing #ai #devops

10 min read

Riley Li

Sep 2

Don't Let a Free LLM Tier Vanish: A Self-Healing Token Budget in 80 Lines

#ai #python #opensource #tutorial

5 min read

Riley Li

Sep 1

Free Models and a Free Server: A 100-Line LLM Failover Proxy

#llm #python #webdev #opensource

5 min read

Riley Li

Aug 31

Free Tokens or Your Own GPU? A Practical Fit-Test for AI Workloads

#ai #llm #selfhosted #productivity

5 min read

Riley Li

Aug 30

Free LLM Tiers vs. Self-Hosted Inference: A Four-Score Script That Picks for You

#ai #llm #productivity #programming

4 min read

Riley Li

Aug 29

Free Models and a Free Server: A 30-Line Budget That Lasts the Month

#ai #llm #productivity #opensource

4 min read

Riley Li

Aug 28

A 60-Line Probe That Decided Between a Free Server and My Own Hardware

#ai #llm #python #serverless

5 min read

Riley Li

Aug 26

Free LLM Tier or Self-Hosted? Three Numbers Decide, Not the Price Tag

#ai #llm #opensource #devops

5 min read

Riley Li

Aug 25

Token Meters for Free LLM Tiers: Three Leaks My 150-Line Logger Caught in an Hour

#ai #python #llm #productivity

7 min read

Riley Li

Aug 25

Debugging a Flaky LLM Pipeline: Timeouts, Truncation, and a 40-Line Probe

#llm #debugging #python #ai

5 min read

Riley Li

Aug 22

Three Nights of Empty LLM Responses — and the Isolation Trick That Finally Caught the Culprit

#ai #debugging #llm #python

4 min read

Riley Li

Aug 14

Minimax H3 Is Having a Moment. My Prompt Set Doesn't Care.

#ai #testing #python #opensource

4 min read

Riley Li

Aug 14

A Credit-Card-Free Triage Bot for Noisy CI Logs

#ai #devops #automation #tutorial

5 min read

Riley Li

Aug 13

Every Week a New Model Is "Cheaper and Better" — Here's the 30-Minute Harness That Settles It

#ai #productivity #testing #tutorial

5 min read

Riley Li

Aug 10

I Stopped Reading Model Release Threads and Built a Release-Day Eval Ritual Instead

#ai #productivity #testing #tooling

5 min read

DEV Community

Badges

Writing Debut

Skills/Languages

Currently learning

Currently hacking on

Available for

Inventory What You Cannot Pin Before You Choose Free Agent Compute

Budget the Agent Loop Before You Choose Shared or Isolated Compute

Score the Host Before You Promote a Vibe-Coded Agent

Choose Agent Compute by Replayable Evals, Not by Spare Capacity

Pass Four Fit Gates Before You Choose Free or Isolated Agent Compute

Who Owns the Failed Agent Run? Score Free Shared Compute Against a Box You Control

Free Compute or Self-Hosted? A Break-Even Guide for Agent Workloads

Rank Agent Runtimes by the Contract You Can Enforce

Invert Non-Negotiable Constraints Before You Accept Free Agent Compute

Keep the Agent Loop Portable, Then Choose Free Shared Compute

Pick Agent Compute by Blast Radius, Not by the Price Tag

Choose Agent Compute by What You Can Falsify

Score the Agent Before You Pick the Endpoint

Score the Lane First: Free Shared Workspace, Paid API, or Self-Hosted

Designing for a 10M-Token Ceiling: A Token-Budget Scheduler for Background LLM Jobs

A Token Bucket for Free-Tier LLM Endpoints: Rate Limiting Without a Fancy Gateway

Keep the Regex Writer Until Shadow Receipts Match

Don't Let a Free LLM Tier Vanish: A Self-Healing Token Budget in 80 Lines

Free Models and a Free Server: A 100-Line LLM Failover Proxy

Free Tokens or Your Own GPU? A Practical Fit-Test for AI Workloads

Free LLM Tiers vs. Self-Hosted Inference: A Four-Score Script That Picks for You

Free Models and a Free Server: A 30-Line Budget That Lasts the Month

A 60-Line Probe That Decided Between a Free Server and My Own Hardware

Free LLM Tier or Self-Hosted? Three Numbers Decide, Not the Price Tag

Token Meters for Free LLM Tiers: Three Leaks My 150-Line Logger Caught in an Hour

Debugging a Flaky LLM Pipeline: Timeouts, Truncation, and a 40-Line Probe

Three Nights of Empty LLM Responses — and the Isolation Trick That Finally Caught the Culprit

Minimax H3 Is Having a Moment. My Prompt Set Doesn't Care.

A Credit-Card-Free Triage Bot for Noisy CI Logs

Every Week a New Model Is "Cheaper and Better" — Here's the 30-Minute Harness That Settles It

I Stopped Reading Model Release Threads and Built a Release-Day Eval Ritual Instead