Opportunity at
InMobiApplied Scientist – User Intelligence
This full-time, on-site Applied Scientist role focuses on user intelligence at InMobi. The team studies model architectures and data to sharpen personalization, with the aim of…
Career Tools
About This Role
This full-time, on-site Applied Scientist role focuses on user intelligence at InMobi. The team studies model architectures and data to sharpen personalization, with the aim of making targeting and pricing systems more effective. Current research examines attention-based designs that combine multiple data types, at a scale the posting puts at trillions of events each day. No office location is named for this role, though the stated salary range applies to…
Job Description
This full-time, on-site Applied Scientist role focuses on user intelligence at InMobi. The team studies model architectures and data to sharpen personalization, with the aim of making targeting and pricing systems more effective. Current research examines attention-based designs that combine multiple data types, at a scale the posting puts at trillions of events each day. No office location is named for this role, though the stated salary range applies to the California and New York offices.
The scientist builds a pipeline of hypotheses, tests each one until it can be accepted or rejected with confidence, and moves the successful ideas into production. Work on attention and post-attention designs is core, and every result has to hold up at the millisecond latencies real-time ads require. The role also refines how models are evaluated, both offline and live, so that gains show up in business results.
Further duties include exploring the wide set of data features across the product portfolio, advising product teams on which valuable data to collect, and contributing to shared model code, training pipelines and documentation. The scientist also tells infrastructure teams what they will need over the medium and long term to run experiments faster. External work such as publishing papers, giving talks and presenting at major conferences is encouraged.
The group includes researchers who have worked in this field for over ten years, and the work depends on close cooperation with engineering and product colleagues. Candidates should show depth in machine learning and deep learning, through peer-reviewed papers at venues such as NeurIPS, KDD or RecSys or through models that have performed well at scale. A PhD is preferred, while a master's degree with a strong industry record is also considered, and near-graduates may apply.
Responsibilities
- Build hypothesis pipelines and evaluate each one through to an accept or reject decision
- Move proven models into production and track their business effect
- Investigate attention and post-attention architectures that meet real-time latency limits
- Strengthen offline and live evaluation so model changes produce measurable results
- Recommend which high-value data sources product teams should collect
- Help maintain shared model code, training pipelines and written docs
- Advise infrastructure teams on needs that will speed up experimentation
- Publish research, give talks and present at major conferences
Requirements
- Proven depth in machine and deep learning, evidenced by peer-reviewed papers or models that succeeded in large production settings
- Working skill with PyTorch or JAX, writing complicated operations in vectorized form
- Strong background in deep recommender models, such as two-tower and attention-driven networks
- Experience with huge distributed data platforms and with production pipelines that feed training and serving
Qualifications
- A doctorate in machine learning or a neighbouring technical discipline is preferred, though a master's with a solid industry record is considered too
- Candidates close to graduation are welcome to apply
- Understanding of the ads ecosystem, including OpenRTB, SSPs, DSPs, ad exchanges and MMPs, is an advantage
Core Skills
Benefits
- Base pay and an ESOP grant where eligible
- Health, dental and eye cover, with an HSA contribution from the employer
- Retirement plan with employer matching
- Paid time off, sick leave, leave for special occasions and company holidays
- Parental leave for both maternity and paternity
- Adjustable hours and a wellness allowance
- Free lunch at the office each day, pet insurance and an employee assistance programme
Frequently Asked Questions
Answers are based only on the employer’s listing; where it doesn’t say, neither do we.
Where is this job based?
The posting does not name an office for this role. Its stated salary range applies to the company's California and New York offices.
What qualifications are preferred?
A doctorate in machine learning or a neighbouring technical discipline is preferred. A master's with a solid industry record is considered too, and near-graduates may apply.
What is the base salary?
The posting gives a fixed base between $182,250 and $218,500 USD for the California and New York offices.
Is the role full-time?
Yes. The listing is full-time and on-site.
Sample Interview Questions
AI-generated questions tailored to this specific role — a preview of the full practice set.