Software Engineer · AI & Agents

Engineer buildingAI agents &fast systems.

Personal site of Khaled Altimany — engineer building AI agents, trading systems, and polished products. Blog, prompt library, and reusable agent skills.

View Work Read Blog Contact

Capabilities

What I work with

Languages

Rust
TypeScript
Python
Go
SQL
Bash

AI & Agents

LLM orchestration
Tool / function calling
RAG
Agent frameworks
Prompt engineering
Evals

Backend

Node.js
Axum / Tokio
FastAPI
PostgreSQL
Redis
gRPC / REST

Frontend

Next.js
React
Tailwind CSS
TypeScript
Framer Motion
Three.js

Infra & Platforms

Cloudflare
Docker
GitHub Actions
Terraform
AWS
Observability

8+: Years building software
40+: Projects shipped
10+: AI agents in production
5: Languages used daily

Selected

Featured work

Agent Orchestration Platform

Tool-calling pipelines, evals, and the infra to run agents reliably.

Rust
AI Agents
Infra

Low-Latency Trading System

Market-data ingestion and execution with deterministic latency budgets.

Rust
Systems
Observability

This Site

A performance-first portfolio combining effects from eight sources.

Next.js
Three.js
Cloudflare

See all work →

Activity

Building in the open

Contributions2,405 in the last year · peak 35 on Oct 9, 2025 · since 2017 · sample data

LessMore

Places I've been

Writing

Latest posts

Building Reliable AI Agents That Don't Fall Over

Hard-won lessons on tool design, retries, and observability for autonomous agents that run unattended in production.

May 18, 2026

Using Notion as a Build-Time CMS for a Static Site

How I wire Notion databases into a Next.js static export at build time — with a graceful fallback to bundled samples when credentials are missing.

Apr 27, 2026

What Trading Systems Taught Me About Software Correctness

Lessons from building automated trading systems where a bug costs real money — idempotency, reconciliation, and trusting nothing.

Mar 9, 2026

Library

Latest prompts

Senior Code Review Partner

A prompt that turns an LLM into a rigorous, kind senior reviewer focused on correctness, edge cases, and clear tradeoffs.

engineering

Force a Spec Before Any Code

A planning prompt that makes the model surface assumptions and a verifiable plan before writing a single line — killing the over-eager-coder failure mode.

workflow

Debug by Hypotheses, Not Guesses

A debugging prompt that forces the model to rank hypotheses and design the cheapest experiment to discriminate between them before touching code.

engineering

Let's build something.

Open to interesting problems in AI agents, systems, and product engineering.

Get in touch More about me