Founding Engineer - AI

LeoForce; Long Beach (CA); On-site; USD 180,000 - 275,000; Full time

Benefits offered by this job

Milestone bonus

Job summary

LeoForce is building a stealth-mode AI hardware startup in Long Beach, assembling a first-engineer team to own end-to-end data-to-system pipelines. You will design data flows, set up local training, and drive model deployment on Jetson-class hardware within tight power and memory budgets.

You will shape the stack, guide architecture choices, and define engineering standards as the company scales. This is a founding role with high ownership and impact.

Qualifications

• Strong Python, and comfortable dropping into C/C++ when runtime demands it.
• Real data engineering experience: pipelines, ETL, schema design for large messy data.
• Hands-on running LLMs or SLMs locally rather than cloud API.
• Database fluency: SQL, PostgreSQL, MySQL, SQLite or MongoDB.
• Linux systems fluency, ARM/embedded Linux.
• Docker and containerized deployment experience.
• End-to-end project ownership without a spec; know when to prototype vs. build.
• Ability to handle sensitive personal data with seriousness and trust.

Responsibilities

• Build the data pipeline from raw capture data into a local database with correct schema.
• Stand up training and dev staging environment and run training on the data.
• Select and deploy a small language model that runs locally on constrained hardware.
• Handle hardware bring-up and the integration layer between the model and device.
• Set engineering standards and assist in hiring the next three engineers.

Skills

Python

C/C++

Data engineering

ETL

SQL

PostgreSQL

MySQL

Linux

ARM

Docker

Embedded Linux

LLMs locally

Tools

Docker

Job description

Job Description

Experience: Senior Level

Salary: $180,000 - $275,000 per year

Long Beach,CA, US

Job Details

You will own the full path from raw data to a working AI system running on constrained hardware. That includes:

• Building the data pipeline from the ground up, moving large volumes of raw capture data into a local database and getting the schema right
• Standing up the training and dev staging environment and running training against that data
• Selecting and deploying a small language model that runs locally on NVIDIA Jetson class hardware, and getting it to perform inside real memory and thermal budgets
• Hardware bring up and the integration layer between the model and the physical device
• Setting engineering standards and helping hire the next three engineers on the team

What we are looking for

• Strong Python, and comfortable dropping into C++ or C when the runtime demands it
• Real data engineering experience, pipelines, ETL, schema design, moving large and messy data reliably
• Hands on running LLMs or SLMs locally rather than through a cloud API. Model selection is a collaborative call and we want your opinion
• Database fluency, SQL, PostgreSQL, MySQL, SQLite or MongoDB. We have not picked one yet, so we care that you have opinions more than which one you have used
• Linux systems fluency, ideally including ARM, ARM64 or embedded Linux
• Docker and containerized deployment
• You have owned a project end to end without a spec, and you know when to prototype versus when to build it properly
• You can handle sensitive personal data with the seriousness it deserves. That trust is the company

Strongly preferred, we do not expect all of these

• NVIDIA Jetson experience, JetPack, TensorRT or CUDA on embedded
• Speech systems, ASR, voice cloning, streaming TTS or real time audio
• Fine tuning, quantization or distillation of small models, synthetic data generation
• Multi camera capture rigs, camera sync, photogrammetry, motion capture or performance capture
• A visual effects, animation or video game background. If you have built 3D or capture pipelines, that crossover is genuinely valuable here
• Computer vision, robotics or holographic imaging

A bit about us:

We are a funded, stealth mode AI hardware startup in Long Beach, and we are hiring our first engineer.

We capture a person in our studio, their voice, their face, how they move, and how they actually think, then compress all of that down into something small enough to run on a device you can hold in your hand with no internet connection. That device then brings them to life on real hardware.

Offline is not a privacy feature we added on. It is the product. Nothing leaves the device. No third party APIs, no cloud inference, no sending anyone's voice out to a vendor. Everything is ours and everything runs local.

We are backed by a single investor with a $20M commitment, our patent is finalized, and our capture studio is already built out with a six camera 360 rig, high definition facial capture, and studio grade audio. What we do not have yet is the engineer who turns all of that captured data into a working system. That is this job.

Why join us?

• You are employee number one. Not early, first. You will pick the stack, set the architecture, and set the engineering standards for everyone we hire after you.
• Nothing is decided yet. No database chosen, no stack locked in, no legacy code to inherit. If you have ever wanted to make the calls instead of living with someone else's, this is that seat.
• Funded and stable. The heavy capital spend is already behind us.
• Milestone bonus tied to launch.
• Work nobody has a playbook for. You would be writing it.