Behavioral AI Solutions Client #1 | Senior Data & AI Scientist

SD Solutions

1d

About The Position

Behavioral AI Solutions is seeking a hands-on Senior Data & AI Scientist to own two behavioral-modeling efforts end to end: a user-conditioned Trader World Model that learns latent representations of traders and reproduces their multi-horizon behavioral distributions, and a large-scale sequence forecaster that predicts users’ next-horizon trading activity from intraday behavior. The role covers data preparation, representation learning, model training, large-scale experimentation, and rigorous evaluation of transformer-based models trained on hundreds of millions of behavioral samples using GPUs.

The position is focused on building, training, and evaluating deep sequence and foundation models—not AI integration or prompt engineering. The scientist will work with psychologists, bioinformaticians, AI engineers, and researchers to translate theory into production-grade AI systems. The company develops proprietary Large Behavioral Models intended to predict human decisions and actions, combining behavioral science, psychology, and machine learning for applications across finance, defense, enterprise decision support, and other industries.

Requirements

• At least 6 years of experience in Data Science, Machine Learning, or AI.
• Hands-on experience training deep sequence models using Transformers or attention, and/or fine-tuning foundation models.
• Strong experience with PyTorch and modern deep learning workflows; Lightning is a plus.
• Experience with distributional and representation learning, including modeling latent structure and full distributions rather than only labels.
• Experience designing, running, and analyzing large-scale machine-learning experiments.
• Demonstrated ownership of the complete modeling lifecycle, from data preparation through training, evaluation, and iteration.
• Expert-level Python.
• Strong understanding of deep learning, representation learning, and statistical modeling.
• Experience with high-dimensional and event-sequence or temporal datasets.
• Ability to design rigorous evaluation methodologies, including chronological or held-out splits, leakage control, purge gaps, and honest baseline comparison.
• Comfort with cloud GPU training and practical failure modes, including throughput bottlenecks, out-of-memory issues, data loading and prefetching, checkpointing and resumption, and cost control.

Nice To Haves

• Experience with zero-inflated, heavy-tailed, or hurdle modeling and calibration for sparse targets.
• Experience with world models or model-based and latent-transition modeling, including JEPA-style approaches.
• Experience with financial, trading, or high-frequency behavioral data.
• Background in behavioral science, psychology, economics, or cognitive modeling.
• Experience with distributed GPU training technologies such as FSDP, DeepSpeed, Accelerate, or DDP, and optimizing large-scale training workloads.
• Experience with LoRA, QLoRA, PEFT, RLHF, or other foundation-model adaptation techniques.
• Publications or open-source contributions related to sequence models or foundation models.

Responsibilities

• Train, fine-tune, and evaluate transformer-based sequence models for behavioral prediction on large-scale trading data.
• Design feature representations and embedding strategies for structured, event-sequence behavioral signals.
• Model full multi-horizon behavioral distributions, including activity, volume, PnL, exposure, and hold time, rather than single-point or label predictions.
• Build and outperform a rigorous baseline ladder; design leakage-safe, chronologically split evaluations that substantiate claims of improvement.
• Optimize large-scale GPU training pipelines on Vertex AI, including throughput, memory, data loading, checkpointing and resumption, and cost.
• Own the modeling lifecycle, from data preparation and representation learning through training, evaluation, and iteration.
• Collaborate with psychologists, bioinformaticians, AI engineers, and researchers to translate theory into production-grade AI systems.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed