A

Accenture

Advises businesses and runs their technology

Agentic Quant Manager - QuantAI, Commodity Markets

Quantitative ResearchPosted on Nov 12
Full-timeLondon, UKNo salary listed

About the Job

Requirements

  • A bachelor's degree in mathematics, statistics, physics, engineering, computer science, economics, finance, operations research, or a related quantitative field is required; an associate degree is acceptable with at least two additional years of directly relevant experience and a strong record of applied quantitative work.
  • At least five years of experience in a quantitative role where models informed trading, hedging, pricing, or risk decisions, such as on a commodity, energy, rates, foreign exchange, credit, equity, options, or crypto desk, or in systematic trading, structuring, or a front-office strategy team.
  • Hands-on experience in at least one of derivatives pricing and hedging, forward-curve or volatility modeling, price or demand forecasting, portfolio or asset optimization, or systematic strategy research.
  • Demonstrated rigorous research practice, including walk-forward or out-of-sample backtests against explicit baselines, leakage control, and honest reporting of results that did not hold up.
  • Hands-on experience formulating and solving mathematical optimization problems, such as linear, mixed-integer, stochastic, or robust optimization, tied to commercial or operational decisions.
  • Expert Python for quantitative work, including tabular data processing, high-performance CPU and GPU compute, optimization solvers, and statistical, machine learning, deep learning, foundation model, generative modeling, and geospatial packages, with Git-based collaboration, tests, and code review.
  • Current daily practice building with coding agents such as Claude Code or Codex, including directing agents to write, test, and refactor research code, correcting their errors, and having built or extended at least one agent workflow, tool, or skill used by other people or agents.
  • Hands-on experience applying modern model classes to quantitative problems, including machine learning, deep learning, foundation models, large and small language models, encoder classifiers, and decision models; must have fine-tuned or adapted at least one and measured whether gains held up out of sample against a classical baseline.
  • Evidence of technical leadership, including setting direction on a problem, leading a team or workstream, and rigorously reviewing others' work, including agent-written code.
  • Evidence of strong communication, including presenting quantitative work to traders, clients, or senior stakeholders and clearly and persuasively explaining model logic and limitations in writing and in person.

Responsibilities

  • Choose appropriate methods across statistics, econometrics, mathematical optimization, scenario modeling, geospatial science, weather and climate models, machine learning, deep learning, foundation model fine-tuning and training, and decision models; connect quantitative work to agent harnesses, reusable agent skills, and Model Context Protocol servers, choosing methods that meet evaluation requirements.
  • Build and extend a forecasting stack combining statistical baselines, fine-tuned time-series foundation models, and multi-model pooling behind Model Context Protocol tools, using leakage-safe walk-forward backtests and probabilistic metrics.
  • Design recursive self-improvement loops where agents propose, test, and keep or revert foundation-model tuning changes using validation data, while reserving untouched test data for final evaluation.
  • Formulate and solve optimization problems that turn forecasts into decisions, including gas-purchase mixed-integer programs and demonstrators for ancillary-service offers, congestion-rights auctions, and margin and collateral planning; integrate them into trader-facing agent workflows.
  • Build and evaluate decision models for text-heavy problems such as counterparty screening and news signals, comparing encoder classifiers, natural language inference classifiers, small language models, and large language models using calibration, human-review deferral, and held-out evaluation.
  • Write quantitative code designed for agent use, with typed inputs, explicit units and trading calendars, reproducible calculations, controlled randomness where needed, and tests that expose errors; package it as Model Context Protocol tools and reusable agent skills.
  • Own the quantitative evaluation gate in agent harnesses, checking agent-written research for leakage, baseline parity, units, calendars, and out-of-sample performance.
  • Run hybrid agentic and human codebase reviews, using review agents to find defects, leakage, and test gaps, then applying human judgment to model logic, market conventions, and readiness to ship.
  • Lead client project teams of domain experts, quants, full-stack engineers, and analysts from problem framing through delivery; work with client trading, origination, and risk teams to turn loosely framed problems into evaluated assets and explain model limitations to senior stakeholders.
  • Lead go-to-market initiatives with practice leadership, including shaping offerings, building demonstrators, contributing to proposals, and presenting to client trading and risk leaders.
  • Coach teammates and set review standards for quantitative work across the team.
  • Advance early priorities such as taking a trading demonstrator from prototype to an evaluated backtest, moving a decision-model prototype from synthetic benchmarks to realistic held-out evaluation, and adding commodity-specific checks to shared research and verification workflows.
  • Work with global clients and colleagues across time zones and travel to client sites when required.

Desired Qualifications

  • Commodity or energy trading experience, especially in power, gas, liquefied natural gas, oil products, freight, or agriculture, including physical constraints such as storage, transport, delivery calendars, and settlement conventions.
  • A master's or PhD in a quantitative field, or a master's in financial engineering or operations research.
  • Experience with foundation-model fine-tuning, including parameter-efficient methods, and foundation-model training on market or operational data, plus hyperparameter search.
  • Experience with scenario modeling, geospatial data, or weather and climate models as inputs to market or operational decisions.
  • Experience building Model Context Protocol servers, agent skills, or evaluation harnesses, or using agent frameworks and software development kits.
  • Experience with model calibration, deferral rules, and benchmark design for large-language-model or classifier systems.
  • Mathematical optimization running in production and used by traders, schedulers, or planners.
  • Experience with recursive self-improvement loops where agents propose, evaluate, and keep or revert changes to model tuning, prompts, or harnesses using validation benchmarks, separate final test data, and safeguards against overfitting and reward hacking.
  • Familiarity with accelerator compute stacks such as NVIDIA CUDA, AMD ROCm, or Apple MLX, including GPU training workflows, model serving, or research infrastructure.
  • Familiarity with cloud platforms such as Amazon Web Services or Microsoft Azure for deploying quantitative workloads, model services, or research infrastructure.
  • Client-facing consulting, forward-deployed, or desk-embedded delivery experience, including leading client project teams or contributing to proposals and business development.

About the Company

Accenture

Advises businesses and runs their technology

Accenture advises large businesses and public agencies on strategy, builds their technology systems and runs ongoing operations. Its teams connect business planning with implementation, covering areas such as cloud migration, cybersecurity and customer experience. Consultants work with clients to redesign processes, while engineers integrate software from technology partners and maintain the resulting systems for multinational organizations. Clients pay for project work or continuing services, including application support and business processes such as finance and customer service. Accenture was founded in 1989 and has its headquarters in Dublin, Ireland.

Company size
10,001+
Company stage
IPO
Headquarters
Dublin, Ireland
Founded
1989

Recently Posted Jobs

Explore more opportunities at Accenture

See more jobs