AI Engineering
Build, evaluate, deploy.
I am not starting from tutorials. I already ship backends, payments, and cloud systems. This page is the AI work that sits on that foundation: assistants, recommendation services, and evaluation you can reproduce.
LLM applications
Assistants, agents, tool calling, and product shells that sit on NestJS APIs with clear auth, billing, and failure modes.
Python + Node split
Node owns product and money paths. Python owns model and recommendation work. Contracts, timeouts, and queues between them.
Evaluation
Rubrics, fixtures, and before/after checks so model changes are measured, not guessed.
Production habits
Docker, AWS, observability, and cost control from the same engineering practice used on payments and media systems.
Flagship systems
Trendibble
Creator-brand campaign platform: rate cards, subscriptions, TrendiMatch agent matching, escrow/wallet flows, and real-time messaging across Expo mobile and Next.js.
NestJS · Agents · Expo
Pace Body Goal
Nutrition and fitness platform: NestJS IAM/meals/subscriptions, FastAPI recommendation service (BMR/TDEE/macros), Expo app, and Next.js admin/kitchen tools. Node + Python split in production.
Python · FastAPI · NestJS
Lumina AI
Personal AI assistant for work and life admin: NestJS API, Expo client, trial/billing, and private context under active build.
AI · LLM · NestJS
Related writings
- How to Evaluate AI-Generated Python Code Safely
Score generated Python with hidden tests, a short-lived sandbox, and a separate check for complexity claims.
- Building an AI Recommendation System with Python
From BMR/TDEE rules to content-based and collaborative filtering in FastAPI: features, cold start, evaluation, and what to keep out of the model service.
- Building an LLM Evaluation Framework in Python
A practical LLM eval harness: benchmark tasks, deterministic checks, rubric scores, and JSONL run reports so prompt changes stop being vibes.
- What LangChain Is (and a Tiny Chain You Can Run)
LangChain as orchestration glue for LLM apps, not magic: prompts, models, parsers, and a runnable mini chain for a fictional FAQ desk.
- Node.js + Python Architecture for AI Applications
Splitting an AI product between a Node.js API and a Python service: who owns what, HTTP vs queues, timeouts, shared contracts, and when not to split.
Contact
Open to full-time AI/ML engineering roles and serious product work where backend, cloud, and model systems have to hold in production.
isaiahdaniel203@gmail.com