You and the process

With a person
Anna, in-house recruiter
9 years hiring engineers
30 minutes with a real recruiter
They read your CV with you, on a call, and say where the offers are being lost.
Didn’t find what you were looking for? Tell us what to build
Point72

Post-Training Research Scientist - Forecasting and Reasoning

  • Office
  • 6+ years

Salary

Not stated

Similar roles pay $140K - 215K a year · our estimate

AI summary

For members

The whole posting in a few lines. Sign up to read it here and on every role you open.

Sign Up to Read

Description

The team builds and applies large language models to forecast outcomes and support decisions in financial markets.

The research examines how models gather useful evidence, make well-calibrated predictions, and update their views as new information arrives.

The work relies on proprietary data, domain knowledge, and live investment use cases to test these abilities.

The data is messy, noisy, and shifts over time. Current focus areas include training environments, reward design, policy optimization, post-training, and evaluation.

The company is a global alternative investment firm that has been investing for more than 30 years. Steven A. Cohen leads it, and it aims to generate returns for investors through fundamental and systematic strategies across asset classes and regions.

Responsibilities

  • Help set the research direction alongside other researchers, covering environment and state construction, data generation and mixing, training, and evaluation, and move methods that work into production.
  • Create post-training approaches for forecasting and reasoning using supervised fine-tuning, reinforcement learning, on-policy distillation, and similar techniques to improve accuracy and calibration.
  • Assemble point-in-time environments from past observations, documents, earlier predictions, and tools, along with tasks that require gathering information, revising beliefs, and forecasting across several time horizons.
  • Design reward models and credit assignment methods that draw on verifiable intermediate results, expert input, preference data, and feedback that arrives late.
  • Work with engineers on synchronous and asynchronous training pipelines running across multi-node GPU clusters, aiming for efficiency and reliability.
  • Set up rigorous evaluations of forecast accuracy, calibration, reasoning quality, robustness, and downstream value. Use temporal holdouts and controlled ablations to separate real generalization from leakage, distribution shift, and evaluator noise.
  • Partner with quantitative researchers to identify useful model capabilities and deploy successful methods into reliable production systems.

Requirements

  • A bachelor's degree or higher in a quantitative or technical field, or equivalent hands-on experience building advanced learning systems.
  • Proven skill in LLM post-training, reinforcement learning, environment design, or preference learning.
  • Preferred: experience working in domains that cannot be fully verified.
  • Hands-on experience with distributed GPU training, including performance tuning, numerical stability, checkpointing, and reproducibility.
  • Strong Python skills and experience with PyTorch or JAX for research and production model training.
  • Familiarity with distributed training and inference frameworks such as FSDP, DeepSpeed, SGLang, Megatron, verl, or Slime.
  • Ability to build rigorous evaluations covering forecasting, calibration, reasoning, robustness, temporal holdouts, and leakage detection.
  • Background in probabilistic forecasting, time-series modeling, asynchronous reinforcement learning, tool-using agents, or production post-training, gained in academia, at frontier AI labs, or on teams building advanced learning systems.
  • A firm commitment to the highest ethical standards.

Benefits

  • Fully paid health care coverage.
  • Generous parental and family leave.
  • Volunteer opportunities.
  • Support for employee-led affinity groups for women, people of color, and the LGBT+ community.
  • Mental and physical wellness programs.
  • Tuition assistance.
  • A 401(k) savings plan with an employer match, plus other benefits.

Where you’d work

From the office

About the company

Point72

  • Industry: FinTech

Office in New York, United States

Also hiring in United States, London, United Kingdom, United Kingdom and 1 more place

Your chances

We've checked whether it's still hiring and how crowded it is.

  • 19 checks run
  • 0 red flags

Still hiring?

13 checks

No red flags

How crowded?

6 checks

No red flags

Fits Me

How well does this role fit you?

Answer a few questions or drop your CV, and every role gets a fit score with the reasons, this one first.

  • Your field
  • Level
  • Stack
  • Work model
  • Salary floor
  • Must-haves
Details10 facts · Role, Location, Compensation, Company
Tech stack
  • Python
  • PyTorch
  • LLMs
Industry
FinTech
Specialty
Data Science
Region
United States
Pay period
Annual
Show 5 more factsShow less

Role

Category
Data & Analytics
Specialty
Data Science
Experience
30+ years
Tech stack
  • Python
  • PyTorch
  • LLMs

Location

Work model
Office
Region
United States
Office
  • New York, United States

Compensation

Salary
Salary by agreement
Pay period
Annual

Company

Industry
FinTech

Something wrong with this vacancy?

Similar vacancies

Share this vacancy

What's wrong with it?

The employer never sees who reported.

Reason