Staff / Principal Machine Learning Engineer, Inworld AI
Serving, UK
$140-200k
+ Equity
More information about location
AI-powered characters for virtual environments
Be an early applicant
AI-powered characters for virtual environments
101-200 employees
Be an early applicant
$140-200k
+ Equity
More information about location
101-200 employees
Company mission
To transform static software into living AI systems that autonomously evolve to better serve their users.
Role
Who you are
- Inference Optimization. Deep understanding of modern serving frameworks and techniques like vLLM or TRT-LLM
- Model Acceleration. Hands-on experience with quantization, distillation, caching strategies , continuous batching, paged attention, and speculative decoding
- High-Performance Systems. Proficiency in C++, CUDA, Rust, or highly optimized Python. You know how to profile code and squeeze every ounce of performance out of NVIDIA GPUs
- Distributed Systems & Scaling. Experience with Kubernetes, Ray, custom load balancing, multi-GPU/multi-node inference, and reliably handling thousands of concurrent connections
- Public work. Non-trivial systems programming projects, open-source contributions to major inference engines, or deep-dive technical write-ups
- Full-cycle ownership. You can take a model from the research team, containerize it, optimize its serving, and ensure it runs reliably in production
- Background. PhD in CS, Physics, Math, or equivalent practical experience building backend or ML systems
- You don’t need a roadmap to start walking; you’re comfortable picking a direction and building the map as you go
- You believe engineering isn't finished until it’s shipped and stable. You have a bias for impact over purely theoretical optimizations
- You don't just ship code; you obsess over the why. You’re the first to question an architecture if you think there’s a better way to solve the core latency or throughput problem
- You aren't satisfied with "the PM said so." You thrive on deep context and want to understand the fundamental logic behind every decision we make
What the job involves
- A year ago, reliably working agentic systems and sub-second multimodal inference at scale barely existed. Nobody has a decade of experience here
- So we're not screening for a resume template — we're looking for strong people from varied backgrounds who learn fast, thrive in ambiguity, and can show us what they've built, broken, and understood
- We hand you unclear problems and expect you to make them clear. We value engineers who say "I don't know yet" and then design the benchmark or prototype that finds out
- We treat performance, latency, and reliability as first-class product features, not a box to check before launch. Impact comes before everything else, though we support sharing work and open-source contributions that move the field forward. Your work should be visible. Flat structure, fast iterations, minimal process theater
Our take
Dialogue writing for video games has traditionally been a matter of branching conversations based on player input, but this is a time-consuming and complicated process. In modern virtual environments which can contain thousands of characters, the need to save on costs means that conversations with many characters can be shallow and repetitive. Inworld AI's solution allows for the creation of unique and flexible interactions for even the most minor characters, based on just a few descriptive sentences.
Since launching, Inworld has quickly grown from an AI-character startup into one of the key players powering consumer AI. Its early work with companies like Xbox, Disney, and NBCUniversal helped provide the value of real-time AI experiences, and now a wide range of media companies, game studios, and fast-growing startups are building on its platform.
Now, with the recent release of Inworld Runtime, the company is taking an even bigger step. By opening up the internal system they originally built to handle huge, real-time consumer workloads, Inworld is positioning Runtime as the go-to backend for any team looking to scale from an idea to millions of users. Its adaptive graphs, automated MLOps, and one-click experiments remove most of the painful engineering work that usually slows teams now. And for Inworld, this turns Runtime into more than just a product, it becomes the core of the company's future.
Steph
Company Specialist at Welcome to the Jungle
Company
Funding (last 2 of 4 rounds)
Aug 2023
$50m
SERIES A
Aug 2022
$50m
SERIES A
Total funding: $117m
Company HQ
Mountain View, CA
Leadership
Ilya Gelfenbeyn
(Co-Founder & Executive Chairman)Previously co-founded API.AI (now known as Google Dialogflow), a developer platform for building conversational experiences, and served as its CEO before Google acquired the company in 2016. The company also built Assistant.ai, the highest-rated independent voice assistant in the world with 40M+ users. Ilya was also the Founding Head of the Google Assistant Investments program at Google.
Michael Ermolenko
(Co-Founder & CTO)Michael has extensive experience in leading AI teams and led Machine Learning Development at API.ai and the Dialogflow NLU/AI team at Google.
Kylan Gibbs
(Co-Founder & CEO)Kylan has a background in turning advanced AI into products that naturally interact with humans. They came from DeepMind, where they led products for Conversational AI and Generative Models. Prior, Kylan worked on enterprise Machine Learning projects at Bain & Company.
Share this job
View 6 more jobs at Inworld AI