Microsoft
Senior Applied Scientist

How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Senior Applied Scientist
# Senior Applied Scientist – Retrieval Technologies (Microsoft 365 Copilot)
About the Role
We are seeking a Senior Applied Scientist with deep expertise in modern retrieval technologies to help advance the capabilities of Microsoft 365 Copilot, with a special focus on Search, Chat, and Agent experiences. This role is based within Copilot and Agents Core (CACore), where cutting-edge generative AI, personalised search, and retrieval systems power M365 Copilot’s intelligence.
In this high-impact, collaborative environment, you will:
- Shape the future of retrieval architectures for Copilot.
- Partner with engineering, product, and platform teams to design and deploy state-of-the-art retrieval systems serving millions of enterprise users daily.
- Influence technical strategy, collaborate across Microsoft Research, Azure AI, and product groups, and deliver AI-powered solutions that drive user productivity.
Microsoft’s mission? Empowering every person and organisation on the planet. Our culture is built on respect, integrity, and accountability, fostering an inclusive and innovative workplace where everyone thrives.
Responsibilities
Core Duties
Design, experiment, and refine retrieval systems to enhance:
- Grounding quality
- Relevance & personalization
- Reasoning capabilities
- Copilot’s Search, Chat, and Agent experiences
Key Focus Areas
Develop expertise in cutting-edge retrieval mechanisms, including:
- Semantic retrieval (e.g., ColBERT-like late-interaction architectures)
- Dense/sparse hybrid retrieval (fine-tuning embeddings, SPLADE)
- Multi-stage ranking optimisation
- Retrieval-Augmented Generation (RAG)
- Personalised context-aware systems
- LLM-integrated retrieval architectures
- Reasoning-aware retrieval & reinforcement learning
Reasons to use Rodeo
I’m in my final year doing Economics and I don’t know whether to apply for grad schemes now or do a masters first. What do you think?
Honest answer — it depends on where you want to end up. A lot of top grad schemes (Big 4, civil service, banking) don’t need a masters. Let’s look at the ones you’d be competitive for now, and we can decide if a masters actually adds anything.
Also worth knowing: most autumn 2026 applications are open now. Timing matters more than you think.
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
Graduate Consultant — 2026 Scheme
Why you're a good match
StrongYour economics background and your summer at a regional bank line up with what PwC looks for on the consulting scheme. Applications close in four weeks.
See breakdownIt searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Why you're a good match
You’ve got the grades and the economics background, and your bank internship is exactly the experience this scheme looks for. Apply soon — deadlines close within the month.
Experience fit
Your summer at the bank plus your econometrics coursework map directly to the day-one responsibilities on this scheme — client modelling, market briefings, and deal support.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
Drive Innovation & Collaboration
- Translate research into scalable, reliable production systems for Copilot’s core experiences.
- Partner with Microsoft Research, Azure AI, and product teams to integrate retrieval advances into production.
- Champion customer impact by addressing end-user pain points (e.g., relevance, grounding precision, freshness).
- Establish evaluation metrics & success frameworks for retrieval quality and system performance.
Leading & Mentorship
- Provide technical leadership, mentoring junior scientists/engineers.
- Influence academic rigor—publishing in top-tier venues (SIGIR, RecSys, WSDM, KDD, ACL, EMNLP) encouraged.
Qualifications
Required
-
Bachelor’s/qualifying degree in:
- Statistics, Econometrics, Computer Science, Electrical/Computer Engineering (or related field).
-
4+ years relevant experience (e.g., research, predictive analytics, retrieval systems) or
-
Master’s degree + 3+ years experience or
-
Doctorate + 1+ year experience or non-degree equivalent credentials.
-
Security clearance compliance: Pass Microsoft’s Cloud Background Check (required upon hire).
Preferred
Note: Production-scale experience is essential.
- Hands-on expertise with:
- Semantic retrieval, hybrid (dense+sparse) models, query understanding, embedding fine-tuning
- SPLADE, RAG, agentic AI architectures, enterprise search systems
- Python proficiency & PyTorch/TensorFlow experience—balanced trial and enterprise-ready scalability.
- A/B testing & offline-online evaluation techniques for retrieval systems.
- Optimisation focus: Latency, scaling, and cost efficiency in cloud infrastructure.
- Publishing record in top-tier conferences or hands-on work in LLM-grounded search.


Get help with your application
Your very own career expert that helps elevate your application to the next level.
Compensation & Benefits
- Location: United Kingdom
- Typical base pay range: £73,800 – £121,300/year (varies by team criteria).
- Comprehensive benefits (flex plans, employee discounts, L&D funding etc.). For full details, see: Microsoft Global UK Pay Policy
- Position open for ongoing applications until filled (minimum pending, 5 days).
Diversity & Inclusion
Microsoft is an equal-opportunity employer and encourages applicants regardless of: Age, ancestry, armed conflict status, citizenship, ethnicity, gender identity, health status, immigration/visa status, marital status, parenthood status, political affiliation, race, religion, veteran/military status, and any other characteristic protected locally, including sexual orientation.
Accommodations for religions/disabilities during the recruitment process? View assistance policies here.
Glassdoor, LinkedIn
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London
Skills
Location