---
title: 'AI Research Engineer at HASH'
canonical: 'https://feeny.ai/job/ai-research-engineer-hash-london-xg7qsm7znf2v'
type: 'job'
last_seen: '2026-09-05'
---

# AI Research Engineer at HASH

- **Company:** HASH
- **Location:** London, United Kingdom
- **Employment:** full-time
- **Work type:** onsite
- **Posted:** 2026-03-25
- **Last confirmed live:** 2026-09-05
- **Apply:** https://jobs.gem.com/hash/am9icG9zdDpPBrWGBH7vUGWp1GDWzFkr

## Job description

## About HASH

At HASH, we're building infrastructure to [solve information failure](https://hash.ai/about/mission) and help everybody make the right decisions. Our open-source platform helps firms integrate both structured and unstructured information into knowledge graphs that support simulating, optimizing and automating processes.

We’re building toward world models from a radically different perspective to most industry labs. If you want to work on ambitious, original research with the chance to influence both scientific direction and the company itself, we’d love to hear from you.

Our founders include Joel Spolsky (former founder of Stack Overflow; acq. $1.8bn, and Trello; acq. $425m), Jude Allred (former CTO of Glitch) and Dei Vilkinsons (former CEO of Soho Strategy).

## About the role

In the last 6 months we've seen significant growth and are now scaling fast. We're hiring exceptional AI researchers in London and Berlin to shape the next generation of foundation models in Europe, both developing new AI models and architectures in-house and post-training existing open-source base models.

This is not a role focused only on applying existing models. You’ll work on both:

- developing new model architectures in-house in support of deterministic simulation, and
- post-training and adapting state-of-the-art open-source base models

## Requirements

- Strong research track record in machine learning, deep learning, or foundation models: evidenced by research publications, open-source contributions, or other evidence of original technical work
- Deep understanding of modern model architectures, training dynamics, inference optimization, scaling, and evaluation: transformer, diffusion, and exotic model architectures, sequence modeling, multimodal learning, and generative modeling
- Hands-on experience training, fine-tuning, or post-training large models: in particular preference optimization, synthetic data, distillation, continual learning, or alignment
- Strong coding and experimental skills: ideally in Python (and/or Rust) and modern ML tooling
- Ability to move fluidly between research ideas and practical implementation: comfort quickly moving from concept to early-proof right through to release and productization
- High agency, strong taste, and excitement about building something category-defining from Europe: we're looking for researchers who are all-in

## What you'll work on

- Research and develop novel model architectures for reasoning, representation learning, and world modeling
- Post-train and adapt open-source foundation models for high-performance real-world use cases
- Design, run, and analyze experiments across training, evaluation, and scaling
- Work across data, systems, and product to turn research into deployed capability
- Contribute to research direction, technical strategy, and long-term model roadmap
- Help define the culture and standards of a high-talent, research-driven founding team

## Why HASH

- Work on first-principles AI research tied to a large and important real-world mission
- Join at a moment of rapid growth, with outsized scope and influence
- Help build a leading frontier AI lab in Europe with global ambition
- Work alongside founders who have previously built and exited businesses worth billions of dollars
- Be part of a team that moves fast, is enthusiastically open source, and is in it for the long-term

## Benefits

Our team have founded and sold billion-euro companies from scratch. We offer leading equity-weighted total compensation, including competitive salaries and tax-advantaged options.

We meet together quarterly in-person for whole-team retreats, provide engineers with generous learning and tech budgets, and recognise and reward output.

This role pays £130,000-£220,000 base/bonus + meaningful equity.

## About HASH

## Company Overview
- **One-liner**: HASH is an open-source platform for decision-making that integrates live data, constructs ontologies, and creates shared understanding through collaborative agent-based simulation and knowledge modeling tools.
- **Entity Type**: Private (Seed stage)
- **Headquarters**: London, United Kingdom
- **Founded**: 2019
- **Founders**: Joel Spolsky and Jude Allred

## Core Business
- Primary industry/industries: Technology, Information and Internet; Business Decision-making Software; Simulation & Modeling
- Target customers: B2B, Enterprise (with a focus on safety-critical supply chain, biopharmaceutical logistics, and complex systems modeling)
- Mission or purpose statement: "Eliminate information failure and help everybody make the right decisions."

## Products & Services
- **[HASH Platform (hash.ai/hash.dev)]**: An open-source, collaborative workspace for structured knowledge that integrates live data, enables ontology construction, and provides fast, distributed agent-based simulation. Combines knowledge capture and modeling tools for complex systems decision-making.
- **Block Protocol (blockprotocol.org)**: An open standard they drive for interoperable, data-driven blocks, contributing to the commons.
- **Open Source Simulation Engine**: Built in Rust/WASM, focused on agent-based simulation and formal verification of distributed systems.

## Market Standing
- **Valuation/Market Cap**: Not disclosed
- **Key Metrics**: Annual Revenue of approximately $1 million; Total Funding of $2.5 million (1 seed round)
- **Notable Investors/Partners**: Zetta Venture Partners (Lead), Root Ventures, with 6 total investors in the seed round. They also work with the Advanced Research + Invention Agency (ARIA) and scientists from the University of Oxford.
- **Growth Signals**: Headcount grew 14.3% YoY (now 9 employees); actively recruiting for new engineering and research roles; conducting high-profile research programs with ARIA and Oxford on safety-critical supply chains; 2,460 LinkedIn followers with 15.6% yearly growth; employer rating of 5.0/5.0 on Wellfound.

## Competitive Advantages
- **Unique Open-Source Approach**: All core tools are open source, building trust and allowing external validation—rare in the decision-modeling space.
- **Technical Moat**: Combines graph-backed world models, formal verification, and agent-based simulation (Rust/WASM) with AI research. Very few companies operate at this specific intersection.
- **High-Impact Focus**: Working on genuinely hard problems (eliminating information failure) with safety-critical applications (supply chains, biopharma logistics), giving them a defensible niche.
- **Elite Talent Sourcing**: Proactively identify and hire exceptional people through their own tooling and community, not just open applications.

## Strategic Focus
- Current priorities: Building a next-generation platform for decision-making via graph-backed world models and AI; expanding into safety-critical supply chain management with government and research partners; advancing their open standard (Block Protocol).
- Direction for growth: Scaling the engineering team (Rust, TypeScript, Python), deepening AI research (graph learning, formal verification), and expanding use cases in complex systems simulation.

## Why Work Here
- **Culture & Work Style**: "We care about output, ownership, and quality." They value raw intelligence, high-energy, and standout performance. The team works in public, contributing heavily to open source.
- **In-Person Focus**: Operates in-person from offices in **London** (HQ) and **Berlin**, with company-wide get-togethers at least twice a year in Europe. Some remote flexibility is available (e.g., Full-Stack Engineer roles can be in-office or remote).
- **Compensation & Perks**: "Leading compensation" with meaningful equity, competitive salaries, an 'unlimited' equipment and learning allowance, and at least 30 days paid time-off each year.
- **Engineering Culture**: Work on hard problems in Rust, TypeScript, and Python, including formal verification, graph learning, and simulation. They emphasize a fast, conviction-based hiring process with decisions made "within days, not weeks."
- **Hiring Process**: A short introductory call followed by a focused work sample or technical deep-dive. Most hires begin with HASH reaching out, but candidates can apply for specific roles.

## Sources
1. [hash.ai](https://hash.ai/careers)
2. [hash.dev](https://hash.dev/careers)
3. [linkedin.com](https://linkedin.com/company/hashintel)
4. [builtin.com](https://builtin.com/company/hash)
5. [jobs.gem.com](https://jobs.gem.com/hash)
6. [blockprotocol.org](https://blockprotocol.org)

## Other roles at HASH

- [Technical Lead](https://feeny.ai/job/technical-lead-hash-london-w4hvjac5jhmm) — London, United Kingdom
- [Product Designer (UX/UI)](https://feeny.ai/job/product-designer-ux-ui-hash-london-j5ms9egvkbjh) — London, United Kingdom
- [Integrations Engineer](https://feeny.ai/job/integrations-engineer-hash-london-mt5wag63x2eb) — London, United Kingdom
- [Forward-Deployed Engineer](https://feeny.ai/job/forward-deployed-engineer-hash-london-b7q0bwqwmrqq) — London, United Kingdom
- [Design Engineer](https://feeny.ai/job/design-engineer-hash-london-gcmsp6c4ndwq) — London, United Kingdom
- [Brand Designer](https://feeny.ai/job/brand-designer-hash-london-1wytwyqmnrpq) — London, United Kingdom
- [Marketing & Communications Lead](https://feeny.ai/job/marketing-communications-lead-hash-london-a2haat5mqgrw) — London, United Kingdom
- [Solutions Lead](https://feeny.ai/job/solutions-lead-hash-london-0tvys1wc1f5j) — London, United Kingdom
- [Outbound Business Development Lead](https://feeny.ai/job/outbound-business-development-lead-hash-london-xk00zm3t32kh) — London, United Kingdom
- [Enterprise Sales Lead](https://feeny.ai/job/enterprise-sales-lead-hash-london-4gzfmtxbxcc6) — London, United Kingdom