San Francisco Bay Area · Enterprise AI

Production AI for the Bay Area companies that aren't AI companies

The Bay Area builds the models. Most of its banks, health systems, biotechs and factories still have to make them pay. I help those teams ship agentic workflows, retrieval systems and private AI infrastructure that move a number they already report on.

Talk about your SF Bay Area team Industries I work with

21.6%Growth in real GDP of the San Francisco metro, 2018 to 2023 (computed from the BEA chart data below) Source

-1.7%Change in annual average nonfarm jobs, 2022 to 2025 (computed from the BLS chart data below) Source

4,630,041Residents of the San Francisco-Oakland-Fremont metro, 2025 Source

San Francisco has an odd problem for an AI consultant to walk into. Everyone has already seen the demo. Your engineers have friends at the labs, your board reads the same headlines you do, and a vendor pitches you an agent every week. What's usually missing isn't awareness. It's a working system inside a regulated business that survives a compliance review, runs on your data, and moves a metric the CFO trusts.

That's the work I do. I'm an independent enterprise AI consultant based in Minnesota, with 20 years across sales, paid acquisition, marketing technology, CRO and analytics, including more than $50M in managed ad spend. I design, build and optimize agentic AI workflows, RAG and knowledge systems, and private AI infrastructure for financial services, healthcare, legal, insurance and manufacturing. Every engagement starts with one KPI, a fixed scope and a fixed price, and you see working software in a demo every week.

The numbers explain why this matters in the Bay Area right now. Real GDP in the San Francisco metro grew 21.6% from 2018 to 2023, while total nonfarm payrolls in 2025 sat 1.7% below their 2022 level. Output is rising faster than headcount. For the operating companies here, the question is no longer whether to use AI but where it takes cost and cycle time out of work that people currently do by hand.

SF Bay Area's industries, and where AI moves the number.

The sectors that anchor the SF Bay Area economy, and the specific work I do for each.

01 / Industry

Financial services and payments

San Francisco is still a headquarters town for money. Wells Fargo's corporate headquarters sits on Montgomery Street, and Visa moved its global headquarters to Mission Rock next to the ballpark. Around them is a deep bench of banks, asset managers and fintech scale-ups that carry heavy regulatory, fraud and servicing workloads.

How I help

I build agentic workflows for the back-office work that eats analyst hours: dispute and chargeback triage, KYC and onboarding document review, policy and procedure Q&A for operations staff, and first drafts of regulatory responses with citations back to source. Each one is scoped to a number such as average handling time, cases closed per analyst, or first-pass accuracy, with audit logs and human sign-off built in from the start.

Enterprise AI for Financial Services

02 / Industry

Healthcare and life sciences

Kaiser Permanente is headquartered in Oakland. Genentech's headquarters is in South San Francisco and Gilead Sciences is based in Foster City. Between integrated care delivery and biopharma R&D, the Bay Area produces an enormous volume of clinical, regulatory and scientific documents that people still read, route and summarize by hand.

How I help

The work here is retrieval-heavy and privacy-bound. I build RAG systems over SOPs, clinical protocols and regulatory submissions so teams get cited answers instead of search results, and document pipelines that extract and check structured data from forms and reports. PHI stays in controlled infrastructure. The KPI is concrete: review turnaround time, time to find the right protocol, or hours of manual abstraction removed per week.

HIPAA-Compliant Enterprise AI for Healthcare

03 / Industry

Advanced manufacturing

The East Bay makes things. Lam Research, which builds semiconductor wafer fabrication equipment, is headquartered in Fremont, and Tesla's Fremont Factory was the company's first vehicle production plant. That means supplier networks, field service organizations and quality teams sitting on years of technical documentation.

How I help

I build knowledge systems that let field engineers and technicians query manuals, service histories and failure reports in plain language, and agents that draft quality and corrective-action paperwork from structured inputs. For plants with IP that can't leave the building, the models run on private or on-prem infrastructure. Targets are things like mean time to repair, first-time fix rate, and engineering hours spent on documentation.

AI for Manufacturing & Supply Chains

04 / Industry

Scale-ups stuck between pilot and production

The Bay Area has more AI pilots per square mile than anywhere I work. Many never leave the sandbox, not because the model is weak but because nobody owned evaluation, cost, security or the change to the actual workflow.

How I help

For growth-stage and enterprise teams with a promising prototype, I run a fixed-scope push to production: an evaluation harness with test sets tied to the business outcome, guardrails and access controls, cost per task measured and cut, and integration into the system people already use. The deliverable is a system in production with a baseline and a measured change against it.

AI Pilot to Production Rescue

SF Bay Area by the numbers.

Public data on the size and direction of the San Francisco economy.

San Francisco metro real GDPReal GDP, San Francisco-Oakland-Hayward MSA, billions of chained 2017 dollars, 2018 to 2023

Show the numbers

Year	$B
2018	560.6
2019	594.3
2020	596.5
2021	662.4
2022	659.3
2023	681.9

Source: U.S. Bureau of Economic Analysis via FRED. Series RGMP41860, millions of chained 2017 dollars, divided by 1,000 and rounded to one decimal. FRED marks this series as discontinued; 2023 is the latest value published under it.

San Francisco metro nonfarm jobsTotal nonfarm employment, annual average, thousands of jobs, 2020 to 2025

Show the numbers

Year	K jobs
2020	2,282.7
2021	2,335.6
2022	2,462.3
2023	2,458.7
2024	2,436.5
2025	2,419.8

Source: U.S. Bureau of Labor Statistics via FRED. Series SMU06418600000000001A, BLS Current Employment Statistics, annual frequency (annual average), not seasonally adjusted, thousands of persons. FRED labels the area San Francisco-Oakland-Hayward, CA (MSA).

No region has a denser AI research base. UC Berkeley's Berkeley Artificial Intelligence Research lab brings together work in computer vision, machine learning, natural language processing, planning and robotics, and just down the peninsula in Silicon Valley, Stanford's Institute for Human-Centered Artificial Intelligence draws faculty from all seven of the university's schools. Many of the companies building frontier models are headquartered in and around San Francisco.

That density cuts both ways for an operating company. Talent and tools are close, but so is the pressure to chase the newest model instead of finishing the system you started. I don't compete with the labs or the platform vendors. I sit on the buyer's side of the table, pick the right model and architecture for a specific workflow, and hold the project to a business result rather than a benchmark.

I work with Bay Area teams remote-first from Minnesota, which keeps the work focused and the cost in the fixed price rather than in travel. When being in the room helps, I fly in: a kickoff to map the workflow and agree the KPI, a working session with the people who do the job today, or a launch week when the system goes live. Between those, you get a demo every week, a shared scorecard against the baseline, and direct access to me rather than a rotating bench. Engagements run on fixed scope and fixed price, and security, data residency and model governance are part of the design, not a phase at the end.

Questions from SF Bay Area teams.

We're surrounded by AI vendors in San Francisco. Why bring in an independent consultant?

Because vendors sell their product and I'm accountable to your number. I pick models, retrieval approaches and infrastructure based on the workflow, your data and your compliance constraints, and I tie the work to a KPI from day one so you can tell whether it paid off.

You're based in Minnesota. Does that work for a Bay Area company?

Yes. Most of the build is remote, with weekly demos and a shared scorecard. I come on-site for the moments that benefit from it, usually kickoff, a workflow workshop with the frontline team, and launch.

We already have a working AI prototype. Can you take it to production?

That's one of the most common engagements. I add an evaluation harness tied to the business outcome, guardrails and access control, cost measurement, and integration into the tools your people already use, then ship it against a fixed scope.

Our data can't go to a public model API. What are the options?

Private deployments in your cloud tenancy, on-prem inference, or a hybrid where sensitive steps run locally and only low-risk steps call external models. I design for the controls your security and compliance teams will ask about, including logging, access and data retention.

Put AI to work in SF Bay Area, tied to a number.

Tell me the process or the problem. You'll get a straight answer on what it's worth and what it would take.

Let's talk

Production AI for the Bay Area companies that aren't AI companies

SF Bay Area's industries, and where AI moves the number.

Financial services and payments

Healthcare and life sciences

Advanced manufacturing

Scale-ups stuck between pilot and production

SF Bay Area by the numbers.

Questions from SF Bay Area teams.

Put AI to work in SF Bay Area, tied to a number.

Other markets I work in.

Sources