# AI consulting and development in Philadelphia, PA > Health systems and universities first, pharmaceutical manufacturing across the New Jersey and Delaware border, and an insurance sector older than either. If you're in one of them, a wrong answer from a model is a patient or a claim, and the system has to be built knowing that. Canonical: https://www.techfabric.com/ai-development-philadelphia Serving: Philadelphia, King of Prussia, Wilmington, Camden, Allentown, Trenton Headquarters: 1530 E Williams Field Rd, Ste 201, Gilbert, AZ 85295. No office in Philadelphia. --- ## Why Philadelphia work is our work An AI pilot that reads patient records or claims can't go to production until it shows the record behind every answer and stops for a person when it should. Our senior engineers build that with your team: retrieval that cites its source, agents that pause for the clinician or adjuster, and evals on every change. Philadelphia's teaching hospitals and the insurers of Center City set that bar, and drugmakers across the river meet it too. ## AI consulting in Philadelphia Philadelphia's AI consulting work starts with patient and claim data, in the hospitals of University City and the insurers on Market Street. Before any model reads a record we agree which data it may see, where a clinician or an adjuster must decide, and how a wrong answer gets caught, then build only the use cases that hold up under that plan. [AI consulting in Phoenix](https://www.techfabric.com/ai-consulting-phoenix) ## Work a buyer here would recognise Patient indexes and claims files are built from records stitched together across sources, and Athlinks needed a pipeline to do that for more than 400 million athlete records. The study shows the multi-tiered serverless architecture built on Temporal for it, with granular control, automated retries and real-time visibility across the whole run. ## Minimum necessary belongs in the platform Philadelphia runs on health care. Penn Medicine, Jefferson and Children's Hospital of Philadelphia are among the region's largest employers, and Independence Blue Cross pays claims from Market Street. HIPAA's minimum-necessary standard covers all of them, so a report, a service account or an agent that reads more patient data than its task needs is a use the rule doesn't permit, whatever the intent behind it. ## Questions ### How do you enforce minimum necessary access under HIPAA? An agent or a service reads through the same grants a person would, resolved at read time. There is no service account with broad access sitting behind the application, because that is the shape that turns an inefficiency into a breach. ### Do you have an office in Philadelphia? The office is in Gilbert, Arizona, with teams in Amsterdam, Dnipro and Hyderabad. Delivery is remote or hybrid, and we travel for discovery, workshops and the opening weeks of a build. ## What ai consulting and development covers ### Agents that hold state A task that runs for three days, calls six systems and waits twice on a person cannot live in a request handler. We build agents on a durable runtime, so a deploy in the middle of a run does not lose the run, and a retry does not send the same email twice. ### Retrieval that cites what it used An answer nobody can trace is an answer nobody can defend. Retrieval is built so every response carries the documents behind it. ### Evaluation, before and after Without a score, every prompt change is a matter of opinion. We put deterministic scorers and model-based judges in front of the thing so a change ships with a number attached rather than with a feeling. ### Human approval where it belongs Some actions should stop and wait for a person. That gate is part of the system rather than a policy document, so an agent that wants to issue a refund pauses, the approval is recorded, and the run picks up where it left off. ### Cost you can see Model spend, retrieval infrastructure and the review time a system still needs are the numbers that decide whether it stays live after the first quarter. They get instrumented at build time, not discovered on an invoice. ### Governed data access An agent reads through the same grants a person would. On Databricks that is Unity Catalog, so a model that tries to reach a table it has no grant for fails the way any principal fails, rather than because the prompt asked it nicely. ## Working with us from Arizona Arizona does not observe daylight saving, so we are two hours behind you from November to March and three hours behind from March to November. That leaves most of your afternoon overlapping our morning, which is usually enough. Most delivery is remote or hybrid. We travel for discovery, workshops and the opening weeks of a build.