Boston · Enterprise AI

Enterprise AI for Boston's money, medicine and research

Boston runs on regulated, document-heavy work: fund administration, claims, clinical care and drug development. I build AI systems that take cost and cycle time out of that work without putting data or compliance at risk. Every engagement is fixed scope, fixed price and tied to a KPI from day one.

Talk about your Boston team Industries I work with

+14.2%Growth in Boston metro real GDP, 2018 to 2023 (computed from the chart data below) Source

5.03MResidents of the Boston-Cambridge-Newton metro, 2025 Source

85,000Employees at Mass General Brigham, the largest private employer in Massachusetts Source

Greater Boston packs an unusual mix into one metro: some of the largest asset managers and custodians in the country, a national insurer, the biggest private employer in Massachusetts running a Harvard-affiliated hospital network, and a biotech cluster around Kendall Square that people call the most innovative square mile on the planet. What these organizations share is a mountain of high-stakes documents and a low tolerance for mistakes.

That is where AI either earns its place or becomes an expensive demo. I design, build and optimize agentic workflows, retrieval systems and private AI infrastructure for exactly this kind of environment. I have spent 20 years across sales, paid acquisition, marketing technology, CRO and analytics, including more than $50M in managed ad spend, so I measure AI the way I measured campaigns: against a number the business already cares about.

Boston's industries, and where AI moves the number.

The sectors that anchor the Boston economy, and the specific work I do for each.

01 / Industry

Asset management and custody

Boston is the financial center of New England. Fidelity Investments is headquartered on Summer Street and State Street Corporation on Congress Street, and the investment operations that surround firms like these run on reconciliations, prospectuses, client reporting and exception queues.

How I help

I build agents that triage trade and reconciliation breaks, draft first-pass client and regulatory reports from source data, and answer operations questions from policy and procedure libraries with citations back to the source. The KPIs are breaks cleared per analyst per day, report turnaround time and the share of answers that pass review without edits.

Enterprise AI for Financial Services

02 / Industry

Hospitals and academic medicine

Mass General Brigham, founded by Massachusetts General Hospital and Brigham and Women's Hospital, reports 85,000 employees and five Harvard-affiliated teaching hospitals. Beth Israel Deaconess Medical Center adds another major academic system in the same city.

How I help

The work here is administrative load, not diagnosis: prior authorization packets, referral intake, coding support and clinician questions against internal protocols. I deploy these on private infrastructure with PHI controls, audit logging and human sign-off, and track minutes saved per case, denial rates and backlog age.

HIPAA-Compliant Enterprise AI for Healthcare

03 / Industry

Property and casualty insurance

Liberty Mutual is headquartered on Berkeley Street in Boston's Back Bay, and the metro's insurance workforce spans underwriting, claims, actuarial and compliance.

How I help

I build claims intake agents that read first notices of loss, adjuster notes and repair estimates, flag missing information and route files by complexity. On the underwriting side, retrieval systems pull guidelines and prior decisions into the submission review. The KPIs are cycle time from FNOL to first contact, touches per claim and leakage on files the model flags.

Enterprise AI for Insurance & Underwriting

04 / Industry

Biotech and life sciences

Boston has become one of the largest biotechnology hubs in the world, and Kendall Square in Cambridge, next to MIT, holds a high concentration of the startups and research sites behind it.

How I help

Life sciences teams drown in protocols, study reports, regulatory submissions and literature. I build document processing and RAG pipelines that extract, compare and summarize these with traceable citations, so medical writing, regulatory and quality teams spend less time searching and more time deciding. The KPIs are hours per submission section, review cycles per document and time to answer a cross-study question.

Intelligent Document Processing & Document AI

Boston by the numbers.

Public data on the size and direction of the Boston economy.

Boston metro real GDPReal GDP of the Boston-Cambridge-Newton, MA-NH MSA, billions of chained 2017 dollars, 2018 to 2023

Show the numbers

Year	$B
2018	451.3
2019	467.4
2020	463.3
2021	495.9
2022	507.8
2023	515.4

Source: U.S. Bureau of Economic Analysis via FRED. FRED series RGMP14460 (millions of chained 2017 dollars), converted to billions and rounded to one decimal. FRED marks this series as discontinued; 2023 is the latest value published.

Boston metro nonfarm jobsTotal nonfarm employment in the Boston-Cambridge-Newton, MA-NH MSA, annual average, thousands, 2020 to 2025

Show the numbers

Year	K jobs
2020	2,542.6
2021	2,624.2
2022	2,718
2023	2,754.8
2024	2,761
2025	2,740.8

Source: U.S. Bureau of Labor Statistics via FRED. MSA-level BLS CES series SMU25144600000000001 (monthly, not seasonally adjusted, thousands). Annual values are the average of the 12 monthly values for each year, computed by us. The NECTA-based series was not used.

Boston's AI bench starts with its universities. MIT and Harvard are both in Cambridge, with Northeastern, Boston University and Tufts across the river, and together they keep a steady flow of engineers, researchers and domain experts moving into local companies. Kendall Square turns that research into startups and corporate research sites, and the hospital systems add one of the largest hospital-based research enterprises in the country: Mass General Brigham reports a $2.7 billion annual research budget. The practical upshot for an enterprise buyer is that talent and ideas are not the bottleneck in Boston. Getting a model safely into a regulated production workflow, with the data, security and evaluation work that requires, usually is.

I am based in Minnesota and work remote-first with Boston teams, which keeps the cost down and the cadence tight: a weekly demo of working software, a shared KPI dashboard and a fixed price agreed before we start. I come on-site when being in the room actually helps, typically for the kickoff workshop where we map the workflow and pick the metric, for a working session with compliance or clinical leadership, and for launch. Most engagements start with one workflow in one department, prove the number, and expand from there.

Questions from Boston teams.

Can you work within HIPAA and our hospital's security review process?

Yes. For clinical and administrative health data I default to private or on-prem deployment, keep PHI out of third-party model training, log every model call, and design for human sign-off. I work through your security questionnaire and BAA requirements at the start, not at the end, so launch does not stall in review.

Our asset management operations are already heavily automated. Where does AI still help?

In the unstructured work that rules-based automation never covered: exceptions, free-text instructions, client correspondence, policy lookups and first drafts of reports. I pick one queue, measure its current throughput and error rate, and build against that number.

We are a Cambridge biotech. Can AI help with regulatory documents without creating a validation problem?

It can, if it is scoped as decision support with citations and human review rather than autonomous authoring. I build retrieval and extraction pipelines where every generated sentence traces back to a source document, and I set up evaluation sets so your quality team can see accuracy before anything goes near a submission.

Do you come to Boston or is everything remote?

Mostly remote, with on-site sessions when they help: kickoffs, stakeholder workshops and launches. The weekly demo cadence keeps things moving between visits.

Put AI to work in Boston, 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

Enterprise AI for Boston's money, medicine and research

Boston's industries, and where AI moves the number.

Asset management and custody

Hospitals and academic medicine

Property and casualty insurance

Biotech and life sciences

Boston by the numbers.

Questions from Boston teams.

Put AI to work in Boston, tied to a number.

Other markets I work in.

Sources