Questions
What firms ask us before they write.
The same questions arrive from a managing partner, a COO and a head of technology, in roughly the same order. They are answered here the way we answer them on a first call, so that the call can start further in.
Where a question deserves a page, the answer points at it. The guides below are the long answers: how adoption actually moves inside an investment firm, why a model misreads a covenant, and what a forward deployed engineer is for.
The long answers
Four guides, written for the question that brought you here.
Each one opens with the search it answers and says in the first screen what we would say in the room.
- How to get a private equity firm to actually use AI
- You have 200 licences and forty people use them. Why utilisation is the wrong number, what adoption looks like inside an investment team, and the order the work goes in. At /guides/ai-adoption-private-equity.
- How to make AI more accurate on investment documents
- Why a model misreads a CIM, a credit agreement or an LP letter, and the five things that cut the errors. At /guides/ai-accuracy-investment-firms.
- AI usage policy for an investment firm
- Why a fourteen-page policy produces the same result as none, and the one page the desk actually reads: what may go in, what may not, and who decides the edge cases. At /guides/ai-usage-policy-investment-firms.
- What a forward deployed engineer does for an investment firm
- The role, where it came from, what it is scoped to, and when to hire one instead of a vendor. At /guides/forward-deployed-engineer.
The three are linked in full under Read next below, and the firm, the sector and the three engagements are on AI consulting in New York.
Questions
The short answers.
Cost, time, data, which model, how it is measured, and what happens on the first call.
What does By The People Technologies do?
We are a New York AI adoption consultancy for financial services firms: private equity, private credit, hedge funds, venture and the allocators who invest in them. We put AI into the hands of the people who do the work, on their own documents, through three engagements: a prompt engineering workshop, a month-to-month AI adoption retainer, and a forward deployed engineer embedded in your team.
What is the Prompt Engineering Workshop?
A session of ninety minutes to two hours for one team, in person at your offices in New York or online, on the documents that team already has open. A deal team works a live CIM, a credit team an agreement, IR the inbox. There is no generic curriculum. Each person leaves having done one piece of their own work correctly, once, with someone watching. Over 100 sessions delivered at more than 50 financial services firms.
What is the AI Adoption Retainer?
A standing month-to-month relationship aimed at one number: how many people in the firm use these tools well. Four things every month: prompting practice across the whole employee base rather than the early adopters, workflow design for the processes where AI earns its place, direct work with the teams who tried it once and stopped, and a plain monthly account of what changed, written so somebody who was not in the room can read it.
What does a forward deployed engineer do?
An engineer embedded in your team, on the client's floor most weeks, scoped to one process with real volume. Discovery on the process as it actually runs, a design with AI at its centre, and a build your own people write under the engineer's guidance, so the process is theirs to run from the first day. Our engineers are Claude Certified Architects, and the skills, agents and workflows they guide are built on Claude.
Do you help with AI governance and usage policy?
Yes, and it is usually the first thing to fix. Most firms have either no written line or a fourteen-page policy nobody opens, and both produce the same result: people assume the answer is no. We write the one page that says what may go into the tools, what may not, and who decides the edge cases, in the words of the desk rather than the vendor's. It sits inside the firm's own data terms with its AI vendor, and the retainer keeps it current as the tools change. The guide at /guides/ai-usage-policy-investment-firms sets out what the page says.
Are you an Anthropic partner?
Yes. By The People Technologies is part of the Claude Partner Network, Anthropic's programme for the firms that deploy Claude with clients. Salesforce is the firm's other partnership. The workshop is taught on whichever tool your firm has licensed; the partnership matters for the forward deployed engineering work, where the build runs on Claude.
What is the AI Maturity session?
Two days on the work your team already has open, spent on the judgment nobody was taught rather than on prompt tricks: what to give a model, what to tell it about the firm's own process, and how to check what comes back. It follows the free AI readiness diagnostic on this site, which reads back where the firm sits before anyone commits to anything.
Which industries do you work in?
Financial services, and nothing else. Private equity, venture capital, private credit, hedge funds, real assets, secondaries, and the allocators, fund administrators and investor relations teams who sit around them. The work crossing those desks falls into four patterns, reading to a decision, assembly, reconciliation and undocumented knowledge, and those four are where AI earns its place.
What does a prompt engineering workshop cost?
It is priced per session, and a session is ninety minutes to two hours for one team on its own documents. We quote it on the first call once we know the team and the material, because the price depends on the preparation the material needs. It is the smallest engagement we sell, and it is sold first because it tells you the most about whether the rest is worth buying.
How long does an engagement run?
A workshop is one session. The retainer is month to month with no fixed term, because the number it is aimed at moves over months and stops moving when the work stops. A forward deployed engineer is scoped to one process and ends when the scope ends.
Do our documents leave the building?
No. Your people work on their own documents inside the tools your firm has already approved, under your firm's own data terms with that vendor. We bring no platform and we keep nothing. The one thing on this site that stores anything is the readiness diagnostic, and the privacy page describes exactly what it keeps.
Which AI model do you use?
Whichever your firm has licensed. Most clients run Claude, ChatGPT or Microsoft Copilot, and the workshop is taught on the tool your people open every morning. We are an Anthropic partner, and the workflows we help clients build run on Claude, but the method is the same on any frontier model and we say so.
Who is in the room for a workshop?
One team at a time, on its own material: deal teams on live transaction documents, credit teams on agreements and compliance reporting, operations, fund administration, IR and finance each on their own. A session that mixes desks on generic examples is the training your people have already had.
How do you measure results?
In business metrics. The retainer is aimed at how many people in the firm use the tools well, by desk, read monthly, which is a different figure from licence utilisation. A workflow is measured by time and error rate on the firm's own volume, before and after. At The Riverside Company, investor-email triage went from days to under five minutes.
Do you work on site in New York?
Yes. The firm is based in New York, the workshop runs in person at your offices or online, and a forward deployed engineer is on the client's floor most weeks. A second engineering seat sits in Tirana, six hours ahead, for the hours New York is not using.
What size of firm do you work with?
Firms where the work has reading in it and the people who do it are reachable, which has ranged from a single deal team to a firm-wide retainer across several hundred people. The engagement is sized to the team rather than to the firm.
Where do your clients come from?
Referrals. Firms come to us because someone they respect in the industry told them to. That is the entire pipeline, and it holds only while the work holds.
What happens on the first call?
You tell us which tools your people have, what you believe they are doing with them, and the one process that costs the most. We tell you what we would look at first, and which of the three engagements fits. If we are the wrong firm for it, we say so on the call.
Is there a way to see where we stand before a call?
The AI readiness diagnostic on this site asks about seven questions and reads back where the firm sits on six dimensions. It is free, it takes a few minutes, and it does not ask for an email address before it shows you the result.
Ask the one that is not here.
One address, read by the people who run the engagements. Tell us the question and roughly what your firm does, and the reply will be the answer rather than a meeting request.
Subject line: A question