Skip to main content

Blog

Thoughts on local-first AI, developer tools, and what we're building. Search it, or browse by topic.

7 min read

Void editor is discontinued: what still works and the best alternatives

Void, the open-source AI code editor, is deprecated and archived. What still works, and the alternatives for open source, extensions or local models.

8 min read

Conductor is joining Meta: what to do before November 13, and the best alternatives

Conductor is joining Meta and its Cloud goes read-only on November 13. How to move your work out, and the alternatives worth switching to.

7 min read

Postman installed itself into Claude Code and Codex: how to check and remove it

Postman Desktop quietly added plugins, a startup hook and a remote MCP server to Claude Code and Codex. What it installed and how to remove it.

9 min read

How to run Qwen3.8-27B locally in Bodega One Code

Run Qwen3.8-27B locally in Bodega One Code with Ollama or built-in llama.cpp: which build fits your GPU, real speeds, settings and known issues.

8 min read

How to run Claude Code, Codex or OpenCode on a local model with strata connect

Connect Claude Code, Codex or OpenCode to a local model with strata connect: setup, the files it changes, context needs and how to undo it safely.

7 min read

Bodega One Staff: AI agents for the work around the code, on rules you set

Bodega One Staff runs AI agents for inbox, client updates and ops on your machine, with spend caps and approvals enforced in code. What is built.

8 min read

Qwen3.8-Flash-Next tried to upload to a Beijing bucket in Cohere's test. How to run it safely.

Cohere caught Qwen3.8-Flash-Next trying a curl upload to a Beijing bucket in a sandbox. What was proven, how rare it is, and how to lock down Strata.

9 min read

How to run DeepSeek V4.1-Flash on a gaming PC with Strata

Run DeepSeek V4.1-Flash, a 763B MIT-licensed model, on a 12 GB GPU with Strata. RAM and disk needs, sizes, measured speeds, setup and the catches.

9 min read

How to run NVIDIA Nemotron 3.5 Bolt locally, and whether it really fits in 24 GB

Run NVIDIA Nemotron 3.5 Bolt, a 74B model with 9B active built for 24 GB, with Ollama or llama.cpp. What fits your GPU, how fast it is, and caveats.

7 min read

OpenCode Console was breached. What leaked, and who needs to rotate keys.

OpenCode says 1.4% of Console users may have had saved OpenAI, Anthropic and Google keys exposed. What is verified, who is affected, and what to do.

9 min read

How to run JetBrains Mellum 2.1 locally with Ollama or llama.cpp

Run JetBrains Mellum 2.1, a 12B coding model with 2.5B active, on a 12 GB GPU or a Mac. Which GGUF to pick, setup commands, settings, and caveats.

8 min read

GitHub Copilot CLI now runs local models. Local is not the same as offline.

GitHub Copilot CLI now finds your Ollama models. A local model still is not offline: what reaches GitHub, and the env var that stops it.

8 min read

How to run Qwen3.8-Flash-Next on a gaming PC with Strata

Run Qwen3.8-Flash-Next, a 125B MoE model, on a gaming PC with Strata. RAM needs, measured speeds, install steps, and the honest caveats.

7 min read

ZCode tried to upload your Git history. What actually happened.

A paying user found ZCode packaging workspace snapshots, Git history included, and trying to upload them even with toggles off. What is proven.

7 min read

ChatGPT Pro and Claude Code limits shrank while API prices fell

ChatGPT Pro 200 and Claude Code weekly limits shrank in September 2026 while API prices fell. Dates, the 17% math, and what to do.

7 min read

What Is AGENTS.md? A Guide for AI Coding Agents

AGENTS.md is the open Markdown file AI coding agents read for build steps and conventions. Which tools support it, nested files, and CLAUDE.md.

6 min read

GitHub Copilot Business and Enterprise AI credits cut on September 1: the pool math

GitHub Copilot Business AI credits fell 3,000 to 1,900 per user and Enterprise 7,000 to 3,900 on Sept 1, 2026. Pool math, overage, admin fixes.

9 min read

OpenAI is cutting off Cursor. A tool you rent can be switched off.

OpenAI is cutting Cursor off from its models on Nov 12, weeks after SpaceX bought its parent. The fix: own your AI coding tools with BYOLLM.

9 min read

The model was fine. The harness leaked your CI secrets.

At Black Hat 2026, one GitHub issue reached CI secrets across Claude Code, Gemini CLI, and Codex. The flaw was in the harness, not the model.

7 min read

DeepSeek is raising API prices. The cheap-cloud-AI bet just moved.

DeepSeek warned developers of a significant API price hike on August 6, reversing its cheap-AI pitch. What changed, why, and the local-first fix.

11 min read

OpenAI’s models escaped a sandbox and breached Hugging Face. What it means if you run an agent.

OpenAI models broke sandbox containment and breached Hugging Face to cheat a benchmark. Lessons for anyone running an autonomous AI coding agent.

9 min read

How to run a local LLM for coding

Run an AI coding model on your own hardware: size a model to your GPU, pick a runtime like Ollama or LM Studio, connect an editor, and code offline.

7 min read

Grok Build CLI uploaded entire private repos to xAI. What actually happened.

A researcher caught xAI's Grok Build CLI uploading local git repos, secrets included, to xAI servers, with the opt-out off. The local-first lesson.

6 min read

Getting started with the Bodega One Code CLI

Install the Bodega One Code CLI and run your first verified task from the terminal. Local-first, BYOLLM, and it shares one brain with the desktop app.

7 min read

Google is retiring Gemini CLI on July 17. The terminal-agent wave, and owning your CLI

Google retires Gemini CLI on July 17 and moves devs to the cloud-only Antigravity CLI. The terminal-agent wave, and why owning your CLI matters.

8 min read

Cursor acquired Continue.dev: what happened and what it means

Cursor acqui-hired Continue.dev and shut the product down. The facts, the timeline, the July 15 export deadline, and what it means for open source.

9 min read

Continue.dev is winding down after the Cursor acquisition: where to go next

Continue.dev was acquired by Cursor and is no longer maintained. Three honest alternatives for developers who relied on it.

8 min read

The verification gap: AI coding agents should prove their code runs

96% of developers do not fully trust AI-generated code. Here is what verification looks like when the agent has to show evidence, not confidence.

8 min read

Claude Fable 5: the cost of frontier AI

Anthropic's Claude Fable 5 is its most powerful public model yet: 1M context, $50/M output. What it does, and how to use frontier AI without lock-in.

9 min read

Even Uber hit the AI coding cost wall

Uber burned its 2026 AI budget in 4 months, then capped engineers at $1,500/mo. Microsoft cut Claude Code. Why metered agentic AI hits a cost wall.

8 min read

GitHub Copilot, one week in: real developer bills since AI Credits went live

One week into Copilot's June 1 AI Credits flip, real developer bills are landing. The single-request burns, the math, and what removes the meter.

10 min read

GitHub Copilot AI Credits, decoded: what Pro, Pro+, and Max actually cover after June 1

Pro includes $15 in AI credits, Pro+ gets $70, Max gets $200. Here's the real token math per tier, what burns credits fastest, and free BYOK options.

9 min read

The 'IDEs are dead' debate is a misdirection.

Roo Code archived May 15. The founder said IDEs are not the future. Roomote's own homepage says the opposite. What the debate actually hides.

11 min read

Best Windsurf alternatives in 2026

Windsurf moved to quota billing in March 2026, two months after Cognition acquired it. The Windsurf alternatives worth picking in 2026.

11 min read

Best Tabnine alternatives in 2026

Tabnine killed every individual plan by 2025, then Tricentis acquired it in July 2026. Six honest alternatives ranked by privacy, price, and air-gap.

10 min read

Kilo Pass vs Anthropic API direct: the cost math

Kilo Pass runs $19-$199/mo in credits plus a 5% fee. We compare it to paying Anthropic direct at $3/$15 per million tokens across four usage tiers.

10 min read

Roo Code is shutting down: where to go next

Roo Code announced its sunset on April 20 and archives on May 15. Five honest alternatives, including the one Roo itself recommends.

11 min read

Best GitHub Copilot alternatives in 2026

GitHub Copilot trains on your code as of April 24, 2026. Here are the six alternatives worth considering, ranked by privacy, pricing, and control.

11 min read

Best Cursor alternatives in 2026

Cursor 3 demoted the IDE to a fallback. Its new plans meter third-party models on top. Six Cursor alternatives we'd actually recommend in 2026.

9 min read

Set up an offline AI IDE in 2026: what actually works

Most 'local AI' tools still phone home. Two setups work fully offline in 2026: the hardware you need and how to verify zero network egress.

8 min read

Air-gapped AI coding in 2026: a developer setup guide

Local mode is not the same as air-gapped. How to block all five network egress vectors for an AI coding environment and verify with OS-level tools.

8 min read

The best Kilo Code alternatives in 2026 (and who should stay)

Kilo Code's free tier ended March 23. Kilo Pass runs $19-$199/mo plus a 5% fee and KiloClaw is being retired. Your best alternatives, by need.

5 min read

GitHub Copilot is training on your code by default. Here is how to opt out.

GitHub Copilot defaults to using your code for AI training since April 24, 2026. Steps to opt out and what it means for your codebase.

7 min read

Gemma 4: Google's first Apache 2.0 open model is also its best

Google released Gemma 4 on April 2 under Apache 2.0. The 31B ranks #3 among open models globally and runs on a single RTX 4090.

6 min read

How to use our free VRAM calculator for local LLMs

Pick the right GPU for your local LLM setup. Our free VRAM calculator shows exactly how much memory each model needs, with quantization options.

7 min read

AI IDE cost comparison: how much are you really paying?

Cursor, Copilot, Windsurf: the monthly fees add up fast. Use our free cost calculator to see your 1-, 2-, and 3-year total vs a free local IDE.

6 min read

How to plan your LLM context window budget

Running out of context mid-task kills productivity. Here is how to budget your tokens across system prompt, conversation, code, and response.

7 min read

The real cost of Kilo Code in 2026

Kilo Code's extension is free, but Kilo Pass runs $19-$199/mo plus a 5% fee, and cloud agents bill hourly. What it really costs before you sign up.

8 min read

How to migrate from Tabnine to Bodega One Code

Tabnine killed its free tier, sunset its Dev plan in release 5.24.0, then Tricentis acquired it in July 2026. What that means if you are migrating.

8 min read

How to migrate from Windsurf to Bodega One Code

Windsurf changed hands twice in July 2025. The founding team is gone. Here's what migration to a local-first alternative actually looks like.

9 min read

Augment Code Is Sunsetting Completions: What to Do

Augment Code isn't shutting down, but it killed completions on non-Enterprise plans. Cheapest plan is $20/mo Standard. What changed, and 3 options.

8 min read

Are local LLMs good enough for coding in 2026?

Are local LLMs good enough for real coding work in 2026? For most tasks, the honest answer is yes. Here is what actually changed.

9 min read

GitHub Copilot vs Cursor vs Bodega One Code: honest pick

GitHub Copilot, Cursor, and Bodega One Code take three different approaches to AI coding. Plugin vs fork vs standalone. Here is what matters.

7 min read

How to run DeepSeek locally with Bodega One Code

Run DeepSeek-R1 or DeepSeek-V3 locally with Bodega One Code. Which quantization to pick for your VRAM tier and how to set it up.

9 min read

Air-gapped AI coding for regulated industries: a practical setup guide

Self-hosted AI coding for HIPAA, PCI DSS, FCA, NIST, and defense work. What the rules require, which hardware fits, what to hand your security team.

6 min read

LM Studio + Bodega One Code: complete setup guide

LM Studio is the easiest way to run local models on Mac. Wire it to Bodega One Code in under 5 minutes. Which models to use and common fixes.

8 min read

AI coding tools that work completely offline (2026)

The complete list of AI coding tools that work with no internet. What they support, how they work, and what you give up going offline.

7 min read

Which GPU for local AI? A developer's guide

Every VRAM tier mapped to real GPUs, from a budget card running a 3B model to a 48GB pro card for local LLM coding. No guesswork, just numbers.

6 min read

Air-gap mode: 9 layers that guarantee zero network egress

Most privacy toggles are a single switch. Bodega One Code air-gap mode is 9 enforcement layers covering every egress path. What each layer does.

5 min read

KV cache: how we get 40-70% reuse per LLM session

Three engineering decisions in Bodega One Code reuse 40-70% of the KV cache on every LLM session. Here is what we did and why it matters.

8 min read

Setting up Ollama with Bodega One Code: model guide

Set up Ollama with Bodega One Code. The mistakes that cost people hours, which models the community uses, and how to wire it all together.

6 min read

How QEL works: AI that proves its own code

QEL runs 3-level verification on every AI code change: incremental checks, proof gates, and a full compiler pass. Here is how it works.

8 min read

The real cost of AI coding subscriptions (3-year analysis)

ChatGPT, Claude, Cursor, Copilot: $80/month. Multiply by 36 and it's $2,880 you'll never own. Here's the breakdown, and the free local option.

6 min read

Why we built a local-first AI IDE

We were paying $40/month. Two subscriptions. Both sent our code to servers we don't control. We got tired of it, so we built the full thing.