---
title: 'AI Agent Researcher at Vast.ai'
canonical: 'https://feeny.ai/job/ai-agent-researcher-vast-ai-san-francisco-6pf6k7faes29'
type: 'job'
last_seen: '2026-10-15'
---

# AI Agent Researcher at Vast.ai

- **Company:** Vast.ai
- **Location:** San Francisco, CA
- **Employment:** full-time
- **Work type:** onsite
- **Posted:** 2026-07-03
- **Last confirmed live:** 2026-10-15
- **Apply:** https://jobs.ashbyhq.com/vastai/eb0150b0-48e2-4e40-9c9f-610306a5ea3d

## Job description

## About Us

[Vast.ai](https://vast.ai)’s cloud powers AI projects and businesses all over the world. We are democratizing and decentralizing AI computing—reshaping our future for the benefit of humanity. We are a growing and highly motivated team dedicated to an ambitious technical plan. Our structure is flat, our ambitions are out‑sized, and leadership is earned by shipping excellence. We seek engineers with strong intrinsic drive, a true passion for advancing the state of the art, and a mix of architecture, coding, and communication skills. LOCATION: On-site at our office in San Francisco or Westwood, Los Angeles.

## About the Role

As an Agent Research Engineer, you will help advance the highest levels of the tech stack to imbue AI systems with true agency. Collaborating directly with our technical founder and diverse team, we will build the next generation of general learning agents. Possessing and maintaining a wide, deep and holistic knowledge base of cutting-edge research is crucial for advancing the mission. Many are called, but few are chosen.

- Full-Time
- On-site at either our SF or LA offices

Tech Stack Python, Engrish, LLMs, ANNs, C++/Cuda Ideal Experience

- Deep expertise in systems engineering across the tech stack
- Strong holistic background in machine learning theory and practice
- Diverse knowledge of neural architectures/circuits: transformers etc
- Published research at top AI conferences

## Key Responsibilities

- Lead cutting-edge research in AI agents, focusing on memory, reliability, and reasoning
- Prototype and evaluate novel state-of-the-art methods/models

Interview Process After submitting your application, our technical team reviews your credentials. If selected, you'll proceed through the following stages:

- Initial screening (virtual 15 minutes)
- Quick dive into Vast, systems and architectures (virtual 30 minutes)
- LLM assisted coding assessment (virtual 1 hour)
- Meet and greet with coding assessment (on-site 2 hours)

Our goal is to complete the interview process in two weeks.

## Benefits

- Comprehensive health, dental, vision, and life insurance
- 401(k) with company match
- Meaningful early-stage equity
- Onsite meals, snacks, and close collaboration with founders/tech leaders
- Ambitious, fast-paced startup culture where initiative is rewarded

## About Vast.ai

## Company Overview
- **One-liner**: Vast.ai operates a decentralized GPU marketplace that enables developers and AI agents to provision and manage compute power across a global network of hardware, offering a cost-effective alternative to hyperscaler clouds.
- **Entity Type**: Private (raised $30M in funding)
- **Headquarters**: Los Angeles, CA (1100 Glendon Ave, STE 1840) and San Francisco, CA (100 1st Street, STE 2250)
- **Founded**: 2016 (incorporated June 28, 2016)
- **Founders**: Jake Cannell (CEO) and Christian Horne

## Core Business
- **Primary industry**: Cloud infrastructure / GPU compute / AI infrastructure
- **Target customers**: AI researchers, ML engineers, AI agent developers, enterprises needing GPU compute for training and inference
- **Mission**: "To organize, optimize, and orient the world's computation."
- **Vision**: "To make life substrate-independent through Vast Artificial Intelligence."

## Products & Services
- **GPU Cloud**: On-demand instances across 20,000+ GPUs in 40+ data centers. Deploy via CLI, SDK, Python API. Per-second billing. Real-time pricing set by supply and demand. [vast.ai](https://vast.ai/)
- **Serverless Inference**: Deploy models as endpoints with automatic GPU optimization, auto-scaling to zero, pay-per-compute-time. [vast.ai](https://vast.ai/)
- **GPU Clusters**: Dedicated multi-node clusters with InfiniBand networking for large-scale training. [vast.ai](https://vast.ai/)
- **API & SDK**: REST API, Python SDK, CLI for programmatic compute provisioning. Agent-native interface for autonomous compute procurement. [vast.ai](https://vast.ai/)

## Market Standing
- **Valuation**: Not publicly available
- **Key Metric**: Total funding $30M (as of 2026, per jobsbyculture.com) [jobsbyculture.com](https://jobsbyculture.com/blog/working-at-vast-2026)
- **Notable Investors/Partners**: Not disclosed in available sources
- **Growth Signals**:
  - 310% year-over-year growth (2024–2025)
  - 20,000+ GPUs, 350+ hosts, 700K+ transactions/month
  - SOC 2 Type I certification achieved in 2024
  - Enterprise and Secure Cloud offerings launched; customers include professional data center partners
  - Headcount grew from fully distributed team to 40+ employees across two offices (LA & SF) [vast.ai/about](https://vast.ai/about)
  - Conflicting reports: jobsbyculture.com cites ~30 employees, while vast.ai/about states "40+ employees"

## Competitive Advantages
- **Decentralized marketplace model**: Taps into underutilized GPU hardware (gaming rigs, mining farms, research labs, small data centers) to offer compute at 3–5x cheaper than AWS, no contracts required.
- **Agent-ready infrastructure**: The same API used by developers is designed for AI agents to autonomously procure and optimize compute – a moat for the upcoming agentic economy.
- **Real-time, transparent pricing**: Prices set by supply and demand, programmatically queryable via API. No hidden costs or enterprise sales friction.
- **Heterogeneous hardware support**: 68+ GPU types across 40+ data centers, enabling users to compare and switch between hardware types easily.
- **SOC 2 certified**: Security and compliance for enterprise workloads.

## Strategic Focus
- **Agentic compute**: Building an "infrastructure layer where AI agents design, procure, and optimize their own compute" [vast.ai](https://vast.ai/)
- **Enterprise expansion**: Growing Secure Cloud (certified data centers) and dedicated cluster products for professional customers.
- **Scaling the network**: Increasing GPU count, host partnerships, and geographic diversity while maintaining real-time pricing and low latency.
- **Open infrastructure**: Keeping compute distributed and independent, countering hyperscaler concentration.

## Why Work Here
- **Culture**: "High level of rigor, precision, and professionalism" – employees are stakeholders who own the impact of their work. Fast-paced startup environment where initiative is rewarded. [vast.ai/about](https://vast.ai/about)
- **Work location**: On-site in Los Angeles (Westwood) or San Francisco (SOMA). Not remote – all current openings require on-site presence.
- **Perks**: Comprehensive health/dental/vision insurance, 401(k) with company match, meaningful early-stage equity, onsite meals and snacks, close collaboration with founders and tech leaders. [vast.ai/jobs](https://vast.ai/jobs)
- **Interview process**: Screened by technical team; stages include a 15-min screening, 45-min deep dive, 1-hour LLM-assisted coding assessment, and 2-hour on-site meet-and-greet. Aim to complete in about one week. [vast.ai/jobs](https://vast.ai/jobs)
- **Salary ranges**: For roles like Senior Infrastructure Engineer and GPU Systems Engineer, posted salary $120K–$180K (may vary by role). [vast.ai/jobs](https://vast.ai/jobs)
- **Engineering culture**: Reports directly to CEO/founder Jake Cannell, a prolific writer on AI and compute scaling theory. Emphasis on systems engineering, GPU optimization, and cutting-edge research.

## Sources
1. [vast.ai/about](https://vast.ai/about)
2. [vast.ai](https://vast.ai/)
3. [vast.ai/jobs](https://vast.ai/jobs)
4. [jobsbyculture.com](https://jobsbyculture.com/blog/working-at-vast-2026)
5. [vast.ai/jobs/apply/systems-gpu-research-engineer](https://vast.ai/jobs/apply/systems-gpu-research-engineer)

## Other roles at Vast.ai

- [Technical Customer Support (L1)](https://feeny.ai/job/technical-customer-support-l1-vast-ai-mexico-w69cjb63gw7b) — Mexico
- [Technical Customer Support (L1)](https://feeny.ai/job/technical-customer-support-l1-vast-ai-vietnam-8gccsn3y9g5a) — Vietnam
- [Head of Sales](https://feeny.ai/job/head-of-sales-vast-ai-los-angeles-yrkth8v89kkh) — Los Angeles, CA
- [Developer Relations Engineer](https://feeny.ai/job/developer-relations-engineer-vast-ai-san-francisco-19dx882ce5x1) — San Francisco, CA
- [Controller](https://feeny.ai/job/controller-vast-ai-los-angeles-0d4jvgnj6v48) — Los Angeles, CA
- [AI, HPC & GPU Infrastructure Support Engineer](https://feeny.ai/job/ai-hpc-gpu-infrastructure-support-engineer-vast-ai-los-angeles-40e39w8ek9p1) — Los Angeles, CA
- [Systems Operations Support Engineer — Linux](https://feeny.ai/job/systems-operations-support-engineer-linux-vast-ai-los-angeles-v3hjj8k7jj4v) — Los Angeles, CA
- [Technical Product Manager, Infrastructure](https://feeny.ai/job/technical-product-manager-infrastructure-vast-ai-los-angeles-b0h8y0t8d4fy) — Los Angeles, CA
- [Technical Support Engineer II (Linux)](https://feeny.ai/job/technical-support-engineer-ii-linux-vast-ai-los-angeles-0h1q79ajj6kc) — Los Angeles, CA
- [Director of Engineering](https://feeny.ai/job/director-of-engineering-vast-ai-san-francisco-q3avgemppyhv) — San Francisco, CA