Andrew Kew - DEV Community

Andrew Kew

Jul 15

GitHub's AI agent can be tricked into leaking private repos via a public Issue

#ai #security #github #developer

3 min read

Andrew Kew

Jul 15

Coding agents can write your integration. They can't run it.

#ai #api #developer #devops

3 min read

Andrew Kew

Jul 15

Cursor has an unpatched 0-day. It's been 7 months.

#ai #security #developer #tooling

3 min read

Andrew Kew

Jul 14

Switching to GPT-5.6 in production: it's a schema, cache, and harness problem — not a model swap

#ai #openai #llm #api

3 min read

Andrew Kew

Jul 14

Claude Code burns 5x more tokens before you type a word. Here's where they go.

#ai #anthropic #llm #devops

1

3 min read

Andrew Kew

Jul 13

GhostCommit: the exploit your AI code reviewer can't see

#security #ai #devops #programming

3 min read

Andrew Kew

Jul 10

Anthropic wants to grade AI jailbreaks like CVEs. Here's the framework.

#ai #anthropic #security #llm

3 min read

Andrew Kew

Jul 10

OpenAI just found ~30% of SWE-Bench Pro is broken — and retracted their own recommendation

#ai #openai #benchmarks #llm

2 min read

Andrew Kew

Jul 10

GPT-5.6 Sol matches Claude Fable 5 intelligence at one third the cost

#ai #openai #llm #api

1

2 min read

Andrew Kew

Jul 9

Local LLMs for agentic coding: a real-world viability report

#ai #llm #programming #machinelearning

3 min read

Andrew Kew

Jul 7

Agents optimizing agents: the wins that stick aren't in the prompt

#ai #llm #agents #machinelearning

2 min read

Andrew Kew

Jul 3

Anthropic just published a jailbreak severity scale. Here's what it means.

#ai #anthropic #security #llm

3 min read

Andrew Kew

Jul 1

Fable is back. But the terms of its return should concern you.

#ai #anthropic #llm #regulation

2 min read

Andrew Kew

Jun 30

Coinbase Cut Its AI Spend in Half Without Throttling Engineers - Here's the Playbook

#ai #llm #cloud #devops

1

2 min read

Andrew Kew

Jun 27

Anthropic, Google, and Microsoft just built a shared security team for open source. AI is why.

#ai #security #opensource #linux

1

3 min read

Andrew Kew

Jun 25

OpenClaw and Hermes agree on what an agent is. They disagree on what controls it.

#ai #agents #llm #opensource

1

3 min read

Andrew Kew

Jun 22

Nvidia wants enterprises to run agents safely. NemoClaw is how.

#ai #agents #nvidia #devops

3 min read

Andrew Kew

Jun 22

'"An LLM and a harness": Nvidia''s simple thesis on what agents actually are'

#ai #llm #agents #nvidia

1

2 min read

Andrew Kew

Jun 21

Fable disappeared overnight. That's the best ad for open-weight AI anyone could have run.

#ai #llm #anthropic #opensource

1

3 min read

Andrew Kew

Jun 20

60–95% fewer tokens in your agent loops, same answers. Meet Headroom.

#ai #llm #agents #api

1

2 min read

Andrew Kew

Jun 17

Your AI Gateway needs guardrails — here's how to add them with AWS Bedrock and Kong

#kubernetes #security #ai #devops

2 min read

Andrew Kew

Jun 17

94% of enterprises still can't make AI work at scale. Scale's new report explains why.

#ai #enterprise #machinelearning #data

2 min read

Andrew Kew

Jun 16

AI is shipping code faster than security was built to handle

#ai #security #devops #appsec

3 min read

Andrew Kew

Jun 16

AWS just made AI bot monetization a WAF setting

#aws #ai #api #cloud

2 min read

Andrew Kew

Jun 15

The US government just recalled an AI model - and a verbal jailbreak claim was enough

#ai #anthropic #llm #government

3 min read

Andrew Kew

Jun 15

Vector Search Got You Started. Production AI Needs Tensors.

#ai #machinelearning #architecture #llm

2 min read

Andrew Kew

Jun 15

Why the "AI replaces engineers" narrative keeps failing the data test

#discuss #ai #programming #machinelearning

1

2

3 min read

Andrew Kew

Jun 15

52,000 packages passed every security check. Chainguard blocked them anyway.

#security #npm #ai #devops

3 min read

Andrew Kew

Jun 15

Xiaomi's MiMo Code gets better as tasks get harder. Here's how.

#ai #llm #devops #machinelearning

3 min read

Andrew Kew

Jun 10

Proxy OpenAI Through Kong AI Gateway on Kubernetes

#kubernetes #kong #ai #devops

2 min read

Andrew Kew

Jun 9

Claude Fable 5 is out: Anthropic's Mythos-class model, minus the wall

#ai #anthropic #llm #api

2 min read

Andrew Kew

Jun 9

Microsoft's npm Packages Got Backdoored. Again. And AI Agents Pulled the Trigger.

#security #devops #npm #kubernetes

2 min read

Andrew Kew

Jun 5

Gemma 4 12B: Google's encoder-free multimodal AI now runs on a laptop

#ai #llm #machinelearning #developer

1

2 min read

Andrew Kew

Jun 2

JetBrains just open-sourced the missing piece of self-hosted AI pipelines

#ai #llm #programming #opensource

3 min read

Andrew Kew

Jun 2

Enterprise AI doesn't need a better model. It needs smarter agent logic.

#ai #agents #llm #machinelearning

2 min read

Andrew Kew

Jun 1

Your DIY platform is automation debt with a better outfit

#devops #kubernetes #cloudnative #platform

3 min read

Andrew Kew

May 31

Your MCP servers can read your SSH keys. Anthropic just fixed that.

#ai #security #api #developer

1

3 min read

Andrew Kew

May 31

Replit agents just got a financial identity — and Visa backed it

#ai #api #webdev #programming

3 min read

Andrew Kew

May 30

"The AI did it" won't save you when EU regulators come knocking

#security #webdev #devops #cloud

3 min read

Andrew Kew

May 29

Mistral acquired an AI physics lab. Here's what they're building.

#ai #machinelearning #engineering #llm

2 min read

Andrew Kew

May 27

AI Security Tools Are Drowning Open Source Maintainers — curl Is the Canary

#ai #security #opensource #devtools

3 min read

Andrew Kew

May 26

Stack Overflow is back to 2008 traffic. The programming book is next.

#discuss #ai #programming #career

2

3 min read

Andrew Kew

May 26

The case for using AI to write better code more slowly

#ai #programming #codereview #llm

1

2 min read

Andrew Kew

May 23

NVIDIA's Nemotron Diffusion: One Model, Three Generation Modes, 6 Faster

#ai #machinelearning #llm #nvidia

3 min read

Andrew Kew

May 22

OpenTelemetry Is Now a CNCF Graduate — and It's Coming for Your AI Stack

#opentelemetry #observability #devops #ai

1

3 min read

Andrew Kew

May 20

Anthropic's MCP tunnels and self-hosted sandboxes: keeping agents inside your perimeter

#ai #anthropic #api #devops

3 min read

Andrew Kew

May 19

Anthropic Acquires Stainless: The SDK Layer Is Now Part of the Platform

#anthropic #ai #api #llm

2 min read

Andrew Kew

May 17

GitHub enters the terminal agent wars with Copilot CLI

#ai #github #developer #devtools

2 min read

Andrew Kew

May 16

AI's energy problem has a software fix. Most teams aren't using it.

#ai #infrastructure #devops #cloud

2 min read

Andrew Kew

May 14

Anthropic packaged Claude for small business — and it runs inside the tools they already use

#ai #anthropic #llm #productivity

2 min read

Andrew Kew

May 13

Your AI agent is the new attack vector. It just wants to help.

#ai #security #llm #agenticai

3 min read

Andrew Kew

May 12

Your agent doesn't remember anything. Persistence isn't memory.

#ai #database #agents #webdev

1

3 min read

Andrew Kew

May 12

Anthropic caught its AI agent blackmailing to survive — here's how it's fixing it

#ai #security #anthropic #llm

3 min read

Andrew Kew

May 12

Your text file is the prompt now: LLM's shebang trick

#ai #llm #cli #developer

2

3 min read

Andrew Kew

May 11

Culture beats tooling: five patterns from enterprises actually scaling AI

#ai #llm #openai #machinelearning

2 min read

Andrew Kew

May 10

Enterprise AI Agents Are Everywhere. The Hard Part Is Trusting Them.

#ai #machinelearning #devops #llm

1

3 min read

Andrew Kew

May 9

Anthropic plugs into SpaceX's 220,000-GPU Colossus — and doubles Claude's rate limits

#ai #anthropic #llm #api

1

3 min read

Andrew Kew

May 7

Atlassian opened its 150-billion-object data graph to any MCP agent

#ai #api #devops #agentops

3 min read

Andrew Kew

May 6

12 million tokens, linear cost: Subquadratic's bet against the attention tax

#ai #llm #api #machinelearning

3 min read

Andrew Kew

May 4

Full agentic coding corrodes the very skills it needs to work

#discuss #ai #programming #productivity

2 min read

Writing Debut

Kong

Want to connect with Andrew Kew?