timi the chef 👨🏾‍🍳

There's a difference between vibe coding and AI assisted development. This is what my current coding workflow looks like. This application processes thousands of $$ of transactions per day so it makes sense that, you can't just merge code into an existing functional codebase without relevant checks. 1. I start with a Plan. I ask Codex to do a full review of the codebase and come up with a plan of what should be done. I don't stop there, I give this same plan to Claude (thanks to Microsoft VS Code split panes). 2. Based on this joint recommendation, I ask Claude to implement, with a caveat: I need to manually authorize a change before it's effected. Claude doesn't like this obviously but hey! 3. I run a Git Fetch in VS Code to see who has changed what, without affecting my local code. If significant changes were made by another developer, I loop back to step 1. If not, I just sync changes and continue from step 2. 4. I commit changes to my local repo by the order of priority defined in step 1. No full head on commits, one after the other. 5. I create a staging environment and run the application to confirm that things are performant. 6. After confirming, I push and sync using VS Code. I'm currently exploring using an agent to perform step 5. There are a few in the market already so I'm spoilt for choice. I'll share my findings regardless. A bit orchestrate-y, I must say.

This would be my 8th year without a CV...

Everyday i find more and more vulnerabilities in tools and products people use everyday, the bar is really low and security needs to be the watch word for very software engineer and more importantly the companies that own these platforms, gone are the days of thinking your code is a black box. It is not, any code you ship to production assume it will be read like plain text to any one with a $20 AI subscription, guard and protect your Users PII, harden your frameworks and secure your keys, the game has changed.

Announcing the Artificial Analysis Coding Agent Index! Our new coding agent benchmarks measure how combinations of agent harnesses and models perform on 3 leading benchmarks, token usage, cost and more When developers use AI to code they’re choosing a model, but also pairing it with a specific harness. It makes sense to benchmark that combination to understand and compare performance. The Artificial Analysis Coding Agent Index includes 3 leading benchmarks that represent a broad spectrum of coding agent use: ➤ SWE-Bench-Pro-Hard-AA, 150 realistic coding tasks that frontier models struggle with, sampled from Scale AI’s SWE-Bench Pro ➤ Terminal-Bench v2, 84 agentic terminal tasks from the Laude Institute and that range from system administration and cryptography to machine learning. 5 tasks were filtered due to environment incompatibility ➤ SWE-Atlas-QnA, 124 technical questions developed by Scale AI about how code behaves, root causes of issues, and more, requiring agents to explore codebases and give text answers Analysis of results: ➤ Opus 4.7 and GPT-5.5 lead the Index: Opus 4.7 in Cursor CLI scores 61, followed closely by GPT-5.5 in Codex and Opus 4.7 in Claude Code at 60. GPT-5.5 in Cursor CLI follows at 58. ➤ Open weights models are competitive, but still trail the leaders: GLM-5.1 in Claude Code is the top open-weight result at 53, followed by Kimi K2.6 and DeepSeek V4 Pro in Claude Code at 50. These are strong results, but still meaningfully behind the top proprietary models. ➤ Gemini 3.1 Pro in Gemini CLI underperforms: Gemini 3.1 Pro in Gemini CLI scores 43, well below where Gemini 3.1 Pro sits on our Intelligence Index, highlighting that Gemini’s performance in Gemini CLI remains a relative weak spot for Google’s offering. ➤ Cost per task (API token pricing) varies >30x: Composer 2 in Cursor CLI is cheapest at $0.07/task, followed by DeepSeek V4 Pro in Claude Code at $0.35/task and Kimi K2.6 in Claude Code at $0.76/task. At the high end, GPT-5.5 in Codex costs $2.21/task, while GLM-5.1 in Claude Code costs $2.26/task. For both models this was contributed to by high token usage, and in GPT-5.5’s case by a relatively higher per token cost. ➤ Token usage varies >3x: GLM-5.1 in Claude Code uses the most tokens at 4.8M/task, followed by Kimi K2.6 at 3.7M/task and DeepSeek V4 Pro at 3.5M/task. GPT-5.5 in Codex uses 2.8M tokens/task, substantially more than Opus 4.7 in Claude Code at 1.7M/task. In GLM-5.1’s case, higher token usage, cost and execution time were partly driven by the model entering loops on some tasks. ➤ Cache hit rates remain high but vary materially: Cache hit rates range from 80% to 96% across combinations. Provider routing, harness prompt structure and cache behavior can materially change the economics of running the same model given cached inputs are typically <50% the API price of regular input tokens. ➤ Time per task varies >7x: Opus 4.7 in Claude Code is fastest at ~6 minutes/task, while Kimi K2.6 in Claude Code is slowest at ~40 minutes/task. This is contributed to by differences in average turns per task, token usage and API serving speed. Opus 4.7 had materially lower amount of turns to complete a task than all other models while Kimi K2.6 had the most. ➤ Cursor made real progress with Composer 2: Composer 2 in Cursor CLI scores 48, near the leading open-weight model results, while being the cheapest combination measured at $0.07/task. Cursor has stated Composer 2 is built from Kimi K2.5, showcasing they have made substantial post-training gains. This is just the start. We are planning to add additional agents (both harnesses and models). Let us know what you would like to see added next.

Need GPU, preferably H100 or NVIDIA L4

Watin concern poor man like me with prenup. All my asset no pass my MacBook, my table tennis bat and my life.