How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Senior AI Engineer
Location: Hybrid (option to work from home for up to 90 days/year,Almost anywhere) Contract Type: Permanent Persistent Detective of AI Systems at Elliptic—Don’t just build, ship: the AI that financial investigators depend on for real-time financial crime detection.
About the Role
Are you passionate about solving technically challenging problems where AI systems directly catch bad actors moving money across blockchains? Are you excited by the art of turning agentic workflows from prototype into robust, scalable production systems?
This is a role for Senior AI Engineers building the backbone of Elliptic’s copilot, an AI product that helps compliance teams trace illicit fund flows, identify patterns, and respond to risk in real-time.
As the hottest AI-driven blockchain intelligence platform, we need engineers who think like investigators, care about engineering excellence, and ship world-class systems that make a tangible impact on preventing financial crime.
Here, you’ll own critical parts of the Agentic stack—focused on LLM-based reasoning, multi-step reasoning, and integrating chain of thought techniques. Your influence will stretch across architecture decisions, workflow design, and mentorship in an evolving AI team.
*We don’t just work on “cool tech”—we build the layers for financial institutions to stay ahead of intelligent threats.
What You’ll Do
You’ll be central to **眣_design and execution_ for Elliptic’s AI systems._ Build the foundations:
-
Build Agentic Systems
- Design and implement uncertified agent-based workflows (e.g., LangChain, LangGraph) solving real investigator pain points in chain scans, rule violations, and fund tracing.
- Focus on endurance, correctness, and cross-cutting complexity, going beyond "demo" to production-ready systems.
-
LLM & Reasoning Frameworks Handle multi-modal reasoning, structured outputs, and context managements to output strong predictive answers on blockchain data. Implement adaptive reasoning to compress thousands of hours of planning or investigation into coherent actionable explanations. Leverage LLM verify-trace-understand (VVU) cycles to detect malicious flows.
-
Back-End Services & Infrastruktur Build high-risk-scoring event-driven pipelines to connect chain analysis, investigator activity, and AI power night-and-day. Backend services & API design (TypeScript, Node.js) for low-latency transactioning data acceleration. Untangle complex call chaining, caching, and zero-pattern flow issues — ensure great infusion between real-time AI systems.
Reasons to use Rodeo
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.
It 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.
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.
-
Impact & Architecture Raise engineering standards via code reviews, architectural decision records, performance trade-offs conservatism, and cost-behavior insulation recommendations (e.g., latency, SLA-driven downsampling).
-
Mentorship & Process Build-Out Coach new engineers while elevating team generative AI practices. Drive engagement with measurement techniques, observability robustness, prompt "detox" reviews, and unwavering safety rollouts.
What You’ll Achieve in the First 6 Months
By Month 6, you should have:
-
Shipped at least one significant AI-driven end-to-end feature, from sketch-stage (e.g., "automate anomaly IDs reporting") to fully reliable in production.
-
Technically steered AI workstreams—architecture, evaluations, and iterative rollouts. Your direction leads—you own responsibility and feedback loops.
-
Set or improved at least one AI engineering practice—examples include:
- Standardise evaluations (fetch-produced scorecards, LLM-as-judge auto-batching, observe throttling).
- Or inventory costs for safe/promised agentic response onset.
-
Mentored juniors/challengers to ding ship mentality (or help them write unit tests, or level-up).
-
Built strong connections with product, data, full-stack, and DevOps—the technical moonshot is a team effort.
Ideal Fit—What We Need You To Have
2. Hard Skills
-
6–8 years’ software engineering experience (including 1–2 years defining/rebuilding LLM-powered backend solutions).
-
Trouble-shoot multi-step API workflows with ease: design, verify, throttle—even debugging enqueue delays at scale.
-
Hands-on experience with LLM frameworks like:
- LangChain
- LangGraph
- Co-pilot plug-ins, Gemma CodeExec
-
Expertise in agentic reasoning tools with human-in-loop patterns:
- Tool integration
- Planning and memory
- Procedural schematised navigation
- Structured output execution
-
Full bash/python + backend stack proficiency in combining: TypeScript/node.js → RPC APIs, AWS EC2, Lambda, S3, step function orchestration + gRPC (or GraphQL)
Get help with your application
Your very own career expert that helps elevate your application to the next level.
-
Evaluations, events across models:
- RAG capabilities
- Error-triggered retry circuits
- Ceasing false generation drift alerts
-
Strong observability stack knowledge:
- Cross-service event-triggered logging
- SQL (Postgres) / NoSQL (MongoDB, Dynamo)
- End-to-end observability with examples
3. Nice to Have (Bonus)
✓ Familiarity with Vector Databases (pgVector, Milvus*) ✓ Scripted Finetuning 일으키기 or open-weight inference pipelines (VLLM, Bitsandbytes acceleration) ✓ MIT-ish decoupled agent adaptations (Multi-Agent-Paradgm not Drawing tool-envs) ✓ Knowledge in Chainোষ Enlightenment frameworks (Shiba, Celestia, etc.) ✓ Infrastructure flexibility and Terraform/K8s implant validation ✓ Experience with devops observability spanning (Lagisty/Sentry, X-Ray/WandB tools)
Your Career & Growth
How We Work
Flexibility: Hybrid model—90 remote working days/year plus £500 setup budget. Team vibe: Divide & "do-ness". Focus on high-task, real-impact shipping (Shell attained outputs, not "+1s").
Learning Broader than Confines
- $1,000 Learning Account—question everything you can make tangible within a domain of global efficacy.
Leave Flexibility
- 25 'net holiday days (plus bank holidays).
- Extra birthday day off (rad)。
- Extended parental leave (definitely paid, 16 weeks non-binary/family supportive).
Benefits That Boost
- Vitalinity Private Health Insurance
- Spin Mental Health Support
- Cycle to Work Scheme (regular pedals included)
- Life Assurance up to 4x your salary (should the worst occur)
Why This Matters
This is an opportunity to expand blockchain investigations into AI-agent world-level impact. If you:
- Stand for engineered-in quality, explain intent over years, and elevate team superებლ semasa,
- Love solving puzzles behind finance, or especially want your tech to fight fraud,
Apply here now. Your skill will be an invaluable fix—step in.
Elliptic’s AI-copoller sees lifeline through you. 🚀
Links: www.elliptic.co/careers (This is not an office meeting.)—Apply asynchronously.
“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