Wall Street trainers charge $25,000 a day to teach banks how to use generative AI

At a glance:

• Ex‑bankers Felipe Sinisterra and Dave Wang are billing up to $25,000 per day for AI‑tool training.
• Their client list so far includes T. Rowe Price, Citigroup and Bank of America.
• They are fully booked for the next two months, creating a two‑month waitlist for banks.

What the trainers do

Felipe Sinisterra and Dave Wang, both former investment bankers, have launched a boutique consultancy called Wall Street Prompt. Their service is simple on the surface: they run live workshops for senior banking staff, showing how to apply commercial generative‑AI models—Anthropic’s Claude, OpenAI’s ChatGPT and Google’s Gemini—to real‑world finance tasks. A Bloomberg‑described session even walked participants through a startup founder’s video pitch using Gemini’s video‑understanding mode.

Why banks are turning to ex‑practitioners

Global banks have poured billions into AI infrastructure, model licences and internal tooling over the past two years. Yet the Bloomberg feature notes that many senior analysts still struggle to translate those tools into deterministic outputs required for earnings interpretation, market‑analysis prompting, due‑diligence synthesis and pitch‑deck review. The trainers fill the gap left by high‑level strategy decks, delivering the granular, hands‑on know‑how that internal teams have not yet mastered.

Pricing and market signalling

The $25,000‑a‑day rate mirrors the quarterly fee generation of a managing director at a large U.S. investment bank, sending a clear signal to procurement that the cost is too small to negotiate. It also exceeds what the big‑four consulting firms charge for comparable AI‑training engagements, reflecting a broader shift toward lean, ex‑practitioner consultancies that undercut traditional McKinsey‑Bain‑BCG pricing models for AI‑specific mandates.

Client roster and booking calendar

To date, the trainers have worked with:

• T. Rowe Price
• Citigroup
• Bank of America The demand is evident: both consultants are booked solid for the next two months, and banks are effectively paying for a two‑month waitlist.

Future outlook for AI‑training services

Model vendors such as Anthropic are moving deeper into financial services—e.g., a Moody’s data partnership and full Microsoft 365 integration announced in early 2026. As providers deliver more plug‑and‑play workflows, the premium for bespoke prompting tutorials may erode. Sinisterra and Wang stay ahead by showcasing live, novel use cases that vendor documentation has not yet covered, but the longevity of that advantage remains uncertain.

Bottom line

For now, the scarcity of expert‑led, hands‑on AI training is a lucrative niche, even if the actual knowledge could be replicated by an analyst with a corporate ChatGPT licence over a weekend. The two‑month waitlist underscores how far banks have to go before generative AI becomes a routine part of their daily workflow.