Aamer Mihaysi - DEV Community

Aamer Mihaysi

Oct 8

How much of your agent's sandbox is actually read-only?

#ai #llm #agents

3 min read

Aamer Mihaysi

Oct 7

Four new models, one interface change: from chat completion to decision

#ai #llm #agents

5 min read

Aamer Mihaysi

Oct 6

Blast radius is the unit of trust

#ai #llm #agents

1

5 min read

Aamer Mihaysi

Oct 5

My World-Model Drift Metric Scored a Stale Checkpoint as Stable

#ai #llm #agents

4 min read

Aamer Mihaysi

Oct 2

Prune tool output by rule, leave the reasoning chain alone

#ai #llm #agents

4 min read

Aamer Mihaysi

Oct 1

Prompt search is a hill-climber, and accuracy is the wrong hill

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 30

Uniform INT8 on Recurrent States Is a Default, Not a Decision

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 29

If every layer prefix is a valid model, why do we still pick a size at deploy time?

#ai #llm #agents

5 min read

Aamer Mihaysi

Sep 24

The agent didn't fail. You never told it what "done" means.

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 23

Containers aren't a sandbox: the Fedora agent incident

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 22

The Windows 11 agent runs in the background, and that's the part nobody's talking about

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 21

The benchmark isn't the problem. Your architecture is.

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 20

The agent has no skin in the game. That's the bug.

#ai #llm #agents

1

2

2 min read

Aamer Mihaysi

Sep 19

Your agent deleted prod because you never let it fail in staging

#ai #llm #agents

1

3 min read

Aamer Mihaysi

Sep 18

The agent runtime is the new IDE. The audit trail is the new GitHub.

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 16

The model that argued with my prompt

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 15

The sandbox can't see what's already in the context

#ai #llm #agents

3 min read

Aamer Mihaysi

Sep 14

Your agent isn't lying. It's optimizing.

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 12

The person on the other end of your agent's output

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 11

You gave your agent maintainer access. That's the bug.

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 10

The Research Bottleneck Isn't Ideas — It's the Missing 40% of the Paper

#ai #llm #agents

3 min read

Aamer Mihaysi

Sep 9

The 30–50% violation rate is a baseline, not a verdict

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 8

The agent that shamed a maintainer wasn't rude. It was optimized.

#ai #llm #agents

1

4 min read

Aamer Mihaysi

Sep 7

The real scandal isn't the benchmark gaming. It's the pretending.

#ai #llm #agents

2 min read

Aamer Mihaysi

Sep 6

Open-source doesn't mean safe: the guardrail check for coding agents

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 4

Sandboxes contain the blast radius, not the consequences

#ai #llm #agents

3 min read

Aamer Mihaysi

Sep 3

Your LLM Judge Can't See What Feedback Did — and It's Costing You

#ai #llm #agents

4 min read

Aamer Mihaysi

Sep 2

Your agent can drop the database. Who approves that?

#ai #llm #agents

4 min read

Aamer Mihaysi

Aug 30

The DN42 agent didn't need a budget. It needed a sense of cost.

#ai #llm #agents

2 min read

Aamer Mihaysi

Aug 29

The skill bottleneck is a myth — your agent needs a memory layer

#ai #llm #agents

1

4 min read

Aamer Mihaysi

Aug 28

SWE-Prime: the pass label is a terrible filter for agent training data

#ai #llm #agents

4 min read

Aamer Mihaysi

Aug 26

The reward function is a policy document

#ai #llm #agents

3 min read

Aamer Mihaysi

Aug 25

Benchmark scores are marketing now

#ai #llm #agents

1

4 min read

Aamer Mihaysi

Aug 24

The proof kernel is the only autonomy you get

#ai #llm #agents

4 min read

Aamer Mihaysi

Aug 23

Your agent's memory doesn't live in the sandbox

#ai #llm #agents

3 min read

Aamer Mihaysi

Aug 22

The matplotlib PR isn't an agent problem. It's a moderation problem.

#ai #llm #agents

2 min read

Aamer Mihaysi

Aug 21

Your agent passes SWE-bench. It still can't do science.

#ai #llm #agents

4 min read

Aamer Mihaysi

Aug 20

The agent wrote a hit piece because you asked it to

#ai #llm #agents

1

2 min read

Aamer Mihaysi

Aug 20

The default is unlimited

#ai #llm #agents

1

2 min read

Aamer Mihaysi

Aug 18

Your OS agent should not have your keys

#ai #llm #agents

4 min read

Aamer Mihaysi

Aug 17

OpenCode vs Cursor vs Copilot: the model was never the bottleneck

#ai #llm #agents

1

4 min read

Aamer Mihaysi

Aug 12

The reward signal is the bottleneck, not the model

#ai #llm #agents

3 min read

Aamer Mihaysi

Aug 11

Selection pressure turns benchmarks into fingerprints

#ai #llm #agents

4 min read

Aamer Mihaysi

Aug 10

Who teaches an agent to take the L?

#ai #llm #agents

1

4 min read

Aamer Mihaysi

Aug 10

Sandboxes don't stop the bill

#ai #llm #agents

4 min read

Aamer Mihaysi

Jul 21

The Career Isn't Eroding. It's Relocating.

#ai #llm #security

1

2 min read

Aamer Mihaysi

Jul 20

Bonsai-27B on a Single 3090: What Works and What Doesn't

#ai #python #devops

1

3 min read

Aamer Mihaysi

Jul 19

By Mid-2027 You'll Train LLMs From Scratch, Not Fine-Tune

#ai #opensource #devops

1

2 min read

Aamer Mihaysi

Jul 18

You Don't Need a Better Agent. You Need a Better Debug Loop.

#ai #machinelearning #programming

1

2 min read

Aamer Mihaysi

Jul 15

Fifty Poisoned Samples Is All It Takes

#ai #machinelearning #datascience

3 min read

Aamer Mihaysi

Jul 14

Small Models That Think Harder Beat Big Models That Sound Confident

#ai #datascience #opensource

2 min read

Aamer Mihaysi

Jul 13

Llamafile vs vLLM: Two Ways to Serve a Local Model, and When Each Makes Sense

#ai #llm #security

1

3 min read

Aamer Mihaysi

Jul 11

Most Evals Measure the Wrong Thing

#ai #opensource #programming

2 min read

Aamer Mihaysi

Jul 10

Agent sandboxing will be the default by mid-2027, and most teams aren't ready

#ai #python #programming

5

1

3 min read

Aamer Mihaysi

Jul 9

GLM-5.2: The 1M-Context Open Model That Actually Ships

#ai #machinelearning #opensource

3 min read

Aamer Mihaysi

Jul 8

Leanstral 1.5: Mistral's Open-Source Bet on Formal Verification Agents

#ai #machinelearning #opensource

3 min read

Aamer Mihaysi

Jul 7

The Reality of 1M Context: Testing Qwythos-9B-Claude-Mythos-5

#ai #machinelearning #opensource

2 min read

Aamer Mihaysi

Jul 3

The Persistence Trap: Why Autonomous Agents are a New Security Nightmare

#ai #machinelearning #opensource

1

3 min read

Aamer Mihaysi

Jul 2

Stop Treating LLM Memory as a Database: The Shift Toward Memory as a Skill

#ai #machinelearning #opensource

1

3 min read

Aamer Mihaysi

Jul 1

Stop Chasing the Hype: Real-World Takeaways from DeepSeek-V4-Pro-DSpark

#ai #machinelearning #opensource

1

2 min read

Writing Debut