Head of Research

Omnifold trains frontier models for forecasting and optimization. Our research team is professors and PhDs from OpenAI, Adept, Google, Stanford, and MIT. We build systems to outperform the state of the art algorithms in prediction, optimization, and control, with a focus on supply chain use cases. We are looking for a leader with the following background:

Must have

Background in ML foundations (i.e. PhD in CS theory, statistics, econometrics, operations research / systems engineering, math, or theoretical physics)
Hands on experience training and deploying production ML models
First-line or second line management experience minimum (remit is to scale a team from 5-20 researchers)

Nice to have

Experience with LLM infrastructure - inference, fine-tuning, RL etc.
Success in environments with quantitive models (quant research fund, ranking, ads)
Startup experience

Location:

San Francisco (in-person, 5 days per week)

Omnifold’s Mission

Every bad forecast has a physical consequence. Unnecessary goods are manufactured, shipped, and stored. Emergency air freight is needed for misallocated products. Poor production planning means workers show up with nothing to do, or work frantic overtime. Inefficiency is everywhere.

Our mission is to eliminate waste and accelerate growth for every company with physical products.

About the company

Omnifold.ai uses artificial intelligence to help large and fast-growing brands and multinational companies manage supply chains and investment decisions. Its AI systems analyze data to predict demand and optimize plans for inventory, production, procurement, logistics, and marketing. The company relies on self-improving AI that learns from new data to improve forecasts and recommendations over time. By giving organizations better visibility and forward-looking guidance, Omnifold.ai aims to reduce waste, lower costs, and improve reliability in supply chains and investment outcomes. The goal is to help clients anticipate future needs, make smarter choices, and execute plans more effectively in a complex global market.

Company Size

11-50

Company Stage

N/A

Total Funding

N/A

Headquarters

N/A

Founded

2024

Get referred to Omnifold

See people who can refer or advise you

Simplify's Take

What believers are saying

Omnifold added OpenAI-style enterprise momentum through June 2026 backing from Kleiner Perkins and Lightspeed.
Case studies show 25% more revenue on hard-to-forecast SKUs and 7.5% inventory-cost reduction.
Hiring for research, applied AI, and sales in September 2026 shows active go-to-market expansion.

What critics are saying

Omnifold is only 2024-founded, so customer retention and durability remain unproven.
Its Bay Area hiring push signals heavy burn before recurring revenue becomes obvious.
Blue Yonder, SAP, and Kinaxis can bundle similar forecasting into incumbent suites.

What makes Omnifold unique

Omnifold raised a $30M Series B on June 26, 2026 at $300M valuation.
Its platform trains customer-specific supply-chain models, not one generic APS forecast.
It reports 30% lower forecast error for a multinational manufacturer versus prior planning software.

Help us improve and share your feedback! Did you find this helpful?

Benefits

Company Equity

Growth & Insights and Company News

Headcount

6 month growth

1 year growth

2 year growth

Omnifold

Mar 6th, 2026

What Not Your Mother's Haircare is Prioritizing in 2026

What Not Your Mother's haircare is prioritizing in 2026. In a recent CGT webinar Eric Falkenmayer, Senior Manager of Demand Planning at Not Your Mother's Hair Care, shared how critical trends in AI are playing out inside a fast-growing CPG brand, and where he sees the biggest opportunities ahead. CGT's latest supply chain report paints a clear picture: consumer goods manufacturers are moving from reactive to proactive stand regarding AI, and AI-powered demand forecasting is at the center of that shift. With 53% of companies planning major AI and machine learning upgrades in 2026, the question is no longer whether to adopt, it's how to apply these tools quickly where they'll drive the most value. From Reactive to Proactive When asked about the single most critical priority for supply chain leaders heading into 2026, Eric was direct: it's the shift from reacting to disruptions to making data-driven decisions ahead of them. That means connecting internal data like shipments, POS, CRM and inventory positioning with external signals and using purpose built AI to plan at a speed and scale. The "Holy Grail" of Demand Forecasting For Eric, the highest-value application of AI isn't strictly incremental accuracy gains on the base business. It's forecasting for innovation: new SKU launches, new customer expansions, and new channels where historical comps are limited. He called it "the holy grail of demand forecasting" and pointed to a recent pilot with Omnifold where forecasts based on their purpose built AI tracked significantly closer to actual shipments than the team's original projections using the same underlying data. Purpose-Built Over General-Purpose A recurring theme was the importance of choosing AI tools designed for the specific problem at hand. Eric drew a sharp distinction: a large language model like ChatGPT is built for conversation, and autonomous driving AI is built to navigate roads safely: you wouldn't swap them. Supply chain forecasting demands the same purpose built approach. Alignment between a tool's design and the problem it solves was a key factor in Not Your Mother's decision to partner with Omnifold. Watch the Full Conversation To hear Eric's full perspective including a detailed walkthrough of Demert's IBP process and a discussion on AI adoption gaps and sustainability visit and subscribe to the Omnifold YouTube channel.