The Firm
Brevan Howard Asset Management is one of the world's leading absolute return / hedge fund managers, overseeing assets on behalf of institutional investors including pension funds, endowments, insurance companies, government agencies, private banks and fund of funds.
Founded in 2002, Brevan Howard currently manages over $35bn, trading predominantly in fixed income and FX markets as well as digital assets through BH Digital. The firm employs over 1,200 personnel worldwide with offices in London, Edinburgh, Jersey, Geneva, New York, Austin, Chicago, Hong Kong, Singapore and the Cayman Islands. Brevan Howard has won several industry awards for excellence in risk management, operational robustness and investment performance.
The Team
The firm is undergoing a period of significant growth and is making targeted investments in its technology team. The successful applicant will join the MBO IT Development team, working in a shared codebase and developing and supporting over 100 applications spanning Trade Repository, P&L, Risk and Client Reporting.
The Opportunity
We are looking for an exceptional full-stack technologist — someone who is equally at home designing distributed systems, owning complex data pipelines and shipping AI-powered tooling. Engineering depth is what matters most here; financial services experience is a strong plus but not a prerequisite. You will be thrown at hard, ambiguous problems, expected to own the architecture end-to-end, and trusted to make the right technical calls with minimal hand-holding.
Main Duties & Responsibilities
Core Engineering Responsibilities
- Architect and own high-throughput, numerically precise calculation engines — covering P&L, returns and accounting measures — with a focus on correctness, testability and long-term maintainability over large, evolving codebases.
- Design scalable reporting infrastructure — defining clean data models, templating layers and delivery mechanisms that allow new report types to be added without architectural surgery.
- Extend a complex, shared Trade Repository system to support new instrument types and asset classes — reasoning carefully about schema evolution, backward compatibility and downstream impact across 100+ dependent applications.
- Design and own end-to-end data platform architecture in AWS and Snowflake — from ingestion and schema design through transformation, lineage tracking and delivery — applying the right patterns for reliability, scalability and auditability at each layer.
- Take ownership of platform performance — profiling slow queries, designing clustering strategies, managing warehouse configurations and building cost governance guardrails as data volumes grow.
- Build robust, observable integrations with external data providers — designing for failure, with schema validation, dead-letter handling, automated reconciliation checks and clear alerting when data quality degrades.
- Work directly with quant researchers and traders to translate ambiguous, complex requirements into well-specified, testable software — pushing back where needed, asking the right questions, and building things that are correct not just functional.
- Own production incidents end-to-end: root-cause analysis, hotfixes under time pressure, post-mortems and systemic remediation — liaising with Platform and SRE teams as needed.
AI-Augmented Development
- Lead adoption and embedding of AI coding tools (Claude, Cursor, GitHub Copilot, Codex) across the team, defining best-practice workflows and prompt engineering standards.
- Use AI assistants to meaningfully compress delivery timelines on new features, bug-fixes and refactoring — while maintaining full code ownership and quality accountability.
- Evaluate emerging AI coding tools continuously; run internal benchmarks and proof-of-concepts to identify what genuinely accelerates output.
- Build agentic coding pipelines where appropriate — e.g. automated code-review bots, AI-driven regression-test generation, AI-assisted documentation.
Knowledge Sharing & Culture
- Coach and upskill team members on effective use of AI coding tools through pair-programming, demos and written guides.
- Document AI-assisted workflows so that productivity gains are institutionalised rather than person-dependent.
- Contribute to the firm's wider AI strategy, working with the CTO and Head of Innovation.
Person Specification
Essential — Engineering Depth
- 5+ years of commercial software engineering in Python, C# or Java — with a demonstrable track record of owning system architecture decisions: decomposing complex domains, defining service boundaries, evolving schemas without breakage, and making trade-offs that hold up over time.
- Proven ability to design distributed systems from first principles — you can whiteboard a solution covering data flow, failure modes, consistency guarantees and operational concerns, and then build it. You have opinions on when to use event-driven vs. request-response, and you can justify them.
- Deep relational database expertise — complex query design, indexing strategies, schema migration under live traffic, stored procedure ownership and query execution plan analysis. MS SQL Server experience preferred; comfort with multiple RDBMS a plus.
- Deep, production-proven experience with AWS — S3, Lambda, EC2, Glue, IAM, EKS — including infrastructure-as-code, cost management and security hardening in a regulated financial environment.
- Expert-level Snowflake: data modelling for financial datasets, query optimisation, Snowpipe, tasks, streams, dynamic tables and role-based access control at scale.
- Demonstrable rigour with financial data: handling floating-point precision, currency conversions, corporate actions, stale/missing prices and full audit trail requirements — not just moving data, but understanding what it means.
- Strong experience with Apache Airflow or equivalent — authoring production DAGs with complex dependencies, SLA monitoring, backfill strategies and Kubernetes-native execution (KubernetesPodOperator).
- Strong engineering hygiene: Git branching strategies, CI/CD pipelines, code review culture, unit and integration testing of numerical logic, and observability tooling (Datadog or equivalent) in production.
- Experience designing and consuming REST or GraphQL web services — including async patterns, pagination, error handling and integration testing against third-party financial APIs.
- A high bar for correctness and reliability — you write tests for numerical logic, you think about edge cases before they hit production, and you treat observability (structured logging, alerting, tracing) as a first-class engineering concern, not an afterthought.
Essential — AI & Tooling
- Proven, hands-on experience shipping production code with AI coding assistants such as Claude, Cursor, GitHub Copilot or OpenAI Codex.
- Demonstrable superuser proficiency: concrete examples where AI tools reduced delivery time materially on non-trivial financial software tasks.
- Strong prompt-engineering skills: ability to decompose complex problems into AI-friendly tasks, validate outputs critically and iterate efficiently.
- Comfortable evaluating AI-generated code for correctness, security vulnerabilities and financial accuracy — never blindly trusting model output.
- Experience building or working with agentic / multi-step AI workflows (e.g. LangChain, Claude tool-use, OpenAI function-calling or similar).
Desirable
- Financial services or buy-side experience is a meaningful plus — familiarity with concepts such as P&L attribution, trade lifecycle or risk sensitivities will accelerate your impact, but we are happy to bring the right engineer up to speed on the domain.
- Experience with dbt or similar data transformation frameworks — defining models, tests and data lineage documentation on top of Snowflake at production scale.
- Experience integrating LLM APIs into production internal tooling — e.g. AI-assisted anomaly detection on financial time series, automated report narration or intelligent reconciliation break triage.
- Prior experience in a hedge fund, investment bank or high-stakes data-intensive environment where the cost of a wrong number is real and immediate.
- Scala or JVM experience — relevant to the firm's existing risk and P&L codebase.
What We Offer
Frontier AI Tools
Full access to Claude Pro, Cursor Pro, GitHub Copilot Enterprise and emerging tools.
Intellectual Environment
Work alongside world-class quants, researchers and traders at one of the most respected macro hedge funds.
High-Impact Problems
Your code runs on systems managing $35bn+ AUM. The problems are hard and the feedback loop is immediate.
Mandate to Innovate
Real authority to shape how the team uses AI — your ideas will be implemented, not filed away.
