Forward Deployed Software Engineer – Tactical Edge AI

About the role

Forward Deployed Software Engineer – Tactical Edge AI

Palantir | Seattle, WA | $145K–$275K USD + RSUs

What happens when an AI agent has no cloud connection, limited compute and absolutely no room for failure?

We’re partnering with Palantir to hire a Forward Deployed Software Engineer for a newly formed team bringing advanced AI directly to the tactical edge.

You’ll build and deploy LLM agents capable of operating on constrained hardware in demanding, high-stakes environments. These systems will interface directly with sensors, effectors, autonomous platforms and drone systems—helping defence customers deploy untethered AI capabilities where conventional cloud infrastructure simply isn’t available.

This isn’t another chatbot project. It’s deeply technical engineering at the intersection of LLMs, edge computing, autonomous systems and mission-critical software.

What you’ll be building

Train and fine-tune LLMs for tactical-edge applications

Develop agentic and multi-agent systems for complex mission scenarios

Optimise models for GPUs, CPUs and specialised hardware accelerators

Connect AI agents with sensors, effectors, radars and autonomous systems

Benchmark model capability, latency, reliability and resource consumption

Build scalable, secure and fail-safe production systems

Work directly with customers to deploy and improve systems in real-world environments

Travel up to 25% to support field deployment and customer missions

What we’re looking for

You’ll bring approximately 5–10 years of post-graduate software engineering experience, with a clear technical spike in building LLM, AI-agent or edge-AI systems.

Ideally, you’ll have:

Hands-on experience developing, training and fine-tuning AI models

Experience building LLM agents through direct API calls or frameworks such as LangChain

Exposure to deploying AI on embedded, edge or compute-constrained hardware

Strong proficiency in Python, C++ and/or Rust

Experience with technologies such as PyTorch, Hugging Face Transformers or vLLM

Familiarity with optimisation tools such as Ollama, Mojo, TVM or TensorFlow Lite

Experience contributing to system architecture, reliability, performance and scaling

A background in defence, aerospace, robotics, drones, autonomous systems, big tech or applied research

Experience working directly with customers or across technical and non-technical teams

A bachelor’s degree or higher in Computer Science, Physics, Mathematics or another quantitative discipline

Experience with defence platforms, UxV systems or other safety- and mission-critical technology would be especially valuable.

Why this opportunity stands out

You’ll join a rapidly growing Palantir team working at the bleeding edge of AI and defence—building systems where software has to perform reliably under genuine operational pressure.

You’ll have serious technical resources behind you, direct access to important customers and the opportunity to actively deploy what you build in the field. The team is moving quickly, tackling problems few engineers ever get the chance to work on and helping define how autonomous AI systems operate beyond the cloud.

The hiring manager describes it as the most exciting work of his career—and believes it may be the most exciting work this team will do in theirs.

Location and package

Location: Seattle, Washington

Working model: Hybrid

Relocation: Candidates must already live in Seattle or be willing to relocate

Travel: Up to 25%

Salary: $145,000–$275,000 USD

Equity: RSUs

Employment: Full-time

Hiring: 1–2 positions available

Important eligibility requirement

This position requires US citizenship and eligibility and willingness to obtain a US security clearance. Candidates with an active clearance are strongly encouraged to apply.

Due to US security-clearance requirements, visa sponsorship is not available and Green Card holders are not eligible for this particular position.

If you’ve built intelligent systems that operate beyond the comfort of the cloud—and want to see your work deployed where reliability genuinely matters—we’d love to hear from you.