Unlock the exact workflows from the channel
Every build-along video on Rakib Jahan | AI Builds comes with the real workflow file — free, no strings, just your email.
I Fine-Tuned an AI ~100x Smaller Than GPT, And It Won
I fine-tuned a 4B parameter model on a single 16GB GPU, then tested it against GPT-5.6 Sol. On a frozen 1,750-document AI-vs-human text detection benchmark, the fine-tuned Qwen model reached 0.915 Macro F1. GPT-5.6 Sol: 0.795 Prompt-only Qwen: 0.541 Fine-tuned Qwen 4B: 0.915 But the interesting part isn't just that the smaller model won. It's why. In this video, I break down when prompting stops being enough, how LoRA and QLoRA make fine-tuning possible on consumer hardware, what actually gets trained, and a few bugs that nearly made the entire experiment useless. #FineTuning #LocalLLM #QLoRA
Get the workflow →Most AI Chatbots Break Here, So I Built a Hybrid Agent
I wanted an AI agent that could handle customers going off-script without handing the entire business process over to an LLM. So I built Peakflow: a hybrid system where the important workflow stays deterministic, but AI can detect an interruption, temporarily route the conversation somewhere else, and then return control to the original flow. In the demo, the customer changes passenger count, meal preference, and travel date midway through a flight booking — and the booking still completes. Peakflow is built on top of Typebot and adds AI interruption events, structured LLM outputs, dynamic cards, OpenRouter support, data tooling, secure credentials, Shopify integrations, and more. Source code: https://github.com/rakibj/peakflow-chat Workf Built on Typebot. Typebot's applicable license still applies.
Get the workflow →I Built an AI System That Works for Almost Any Contractor Business
I built an AI operations system that can take a customer request, understand the problem, find the right contractor, handle rejections, collect a quote and ETA, approve the job, and keep the customer updated, automatically. I use property maintenance as the example, but the same architecture can work for plumbing, HVAC, electrical, cleaning, landscaping, appliance repair, and many other contractor or field-service businesses. The system combines AI reasoning with deterministic automation: AI handles messy human communication, while rules handle things like contractor selection, approval limits, retries, and job state.
Get the workflow →AI Employee That Answers Calls & Sends Checkouts on WhatsApp
AI Employee That Answers Calls & Sends Checkouts on WhatsApp
Get the workflow →