Senior AI Engineer (AI Automation, Claude Code, Full Stack)

Samuels Venmoney Estates; United States; Remote; USD 32,000 - 63,000; Full time

Benefits offered by this job

Fully remote

Budget for models and tools

Ownership from idea to live system

Direct founder collaboration

Job summary

Samuels Venmoney Estates seeks an AI expert to design and deliver end-to-end automated solutions, dashboards, and AI-powered features. You will work from concept to live system, ensuring reliability, guardrails, and cost control across communications, document processing, and internal tools.

You will collaborate with the founder, build front-end and back-end capabilities, and own deployment and monitoring from day one.

Qualifications

• 4+ years building software, with at least 2 years shipping LLM features and automations that real users rely on.

Responsibilities

• Design and build AI automations that reduce manual work across email, messaging, and document workflows.

Skills

AI/ML engineering

LLM feature shipping

Full stack development

API integration

OpenAI/Claude APIs

n8n/Make automation

Front-end: React/Next.js

Back-end: Node.js/Python

Tools

Claude Code

OpenAI/Claude APIs

n8n

Make

Docker

AWS

Job description

The role

You are our AI expert. We hand you a problem, and you solve it with AI: an automation that removes manual work, an AI feature that does a job properly, or a dashboard the team runs on. You design it with Claude and ChatGPT, build the back end and the front end with Claude Code, and own it from the first conversation with the founder to a live, monitored system. A voice note in the morning becomes a short spec by lunch and a working first version by the end of the week. Then you harden it.

What you will build

• AI automations that take work off people: email and WhatsApp triage, lead follow-up, document processing, reports, and hand-offs between Gmail, Google Drive, ClickUp, Shopify, WhatsApp and SMS. n8n or Make where they fit, code where they don't.
• AI agents that use tools and MCP servers to take actions, with guardrails, retries, logs and human approval where it matters.
• Transcripts done properly: calls, meetings and voice notes turned into accurate transcripts with speakers and timestamps, then summaries, action items and tasks, automatically.
• AI features inside products: extraction, classification, summaries and search over company documents, with evaluations so quality is measured, not guessed.
• Dashboards and internal tools to run it all: admin panels, reporting dashboards for our team and our clients, and operations screens, front end and back end.

What we expect

• You are an AI expert, not a prompt writer. 4+ years building software, with at least 2 years shipping LLM features and automations that real users rely on. Demos and notebooks do not count.
• Deep, hands‑on use of the OpenAI and Claude APIs: tool use, structured outputs, streaming, prompt caching, evaluations and cost control. You know which model fits which job and what it will cost.
• Automations that hold up: n8n, Make or code, with webhooks, queues, retries, monitoring and error alerts, so they keep running when nobody is watching.
• Full stack: you build web apps end to end, with a React or Next.js front end, a Node.js or Python back end, a database and a deployment you own.
• You design with AI. You use Claude, ChatGPT, v0 or similar to turn an idea into a clean, usable screen without waiting for a designer, and you can tell when a screen is wrong.
• You build with AI coding tools every day, ideally Claude Code, which most of our tools are built with. You read and test what they write.
• You solve problems. When a brief is vague, you ask the questions that matter and then decide. When something breaks, you find the cause instead of patching around it.
• Speed with judgement: a working first version fast, then tests, logging and error handling before it is called done.
• You document what you build, so the team can run it without you.
• Care with personal data. You keep it out of prompts, logs and third‑party tools unless it is allowed there. Some of our work is healthcare.
• Clear written English. You explain trade‑offs to a non‑technical founder in plain words.

Good to have

Speech‑to‑text in production with Whisper, Deepgram or AssemblyAI. RAG and vector databases. Agent frameworks such as LangGraph or the OpenAI and Claude agent SDKs. Voice AI platforms such as Retell, Vapi or ElevenLabs. Shopify or ClickUp APIs. Chart libraries such as Recharts. Docker and AWS. Arabic.

Why join

• You work directly with the founder, and what you build goes live within days.
• Your work runs across several businesses at once: e‑commerce brands, a voice AI product and a healthcare platform.
• A budget for the models and tools you need.
• Fully remote. You own what you build, from idea to live system.

How we work

• Full time, remote, anywhere in India. 9:00 to 18:00 IST, Monday to Friday.
• You report to the General Manager. Work is briefed by voice note or one page, and you confirm a short spec before you build.
• A written update every day, and a Friday demo of what shipped.