# AI consulting and development in Houston, TX > Energy and petrochemicals on one side, the Texas Medical Center on the other, and a long tail of companies serving both. If you run one of them, you generate instrumented data at a volume where an AI answer without its source is unusable, and a regulator will eventually ask how a number was reached. Canonical: https://www.techfabric.com/ai-development-houston Serving: Houston, Sugar Land, The Woodlands, Pasadena, Galveston, Baytown Headquarters: 1530 E Williams Field Rd, Ste 201, Gilbert, AZ 85295. No office in Houston. --- ## Why Houston work is our work Your AI pilot gives answers nobody on the plant floor or the ward will act on, because it can't show the sensor reading or record behind them. Our senior engineers take it to production with retrieval that cites sources at volume, agents that stop for a person where it matters, and evals that score every change. Energy and the Texas Medical Center are Houston's biggest cases, and a port operator or builder needs the same. ## AI consulting in Houston In Houston, AI consulting starts with instrumented data and the regulator behind it. Energy operators and the Texas Medical Center both need answers that cite the reading or the record, so we assess source quality and citation needs first and recommend use cases in the order the data can support. [AI consulting in Phoenix](https://www.techfabric.com/ai-consulting-phoenix) ## Work a buyer here would recognise Sensor histories and clinical records come from systems that never shared a key, and the pipeline built for Athlinks deals with that shape, stitching more than 400 million athlete records from disparate sources. It runs on Temporal with automated retries and real-time visibility, so a failed step reruns and the run stays observable. ## Two regulators can share one platform Two of Houston's large producers of data answer to different rulebooks. A pipeline operator on the Ship Channel keeps integrity records that PHMSA audits under 49 CFR Parts 192 and 195. A sponsor running trials in the Texas Medical Center keeps electronic records under the FDA's 21 CFR Part 11, with HIPAA over the patient data, and the service firms working for both hold records under each. ## Questions ### We have both energy operations and clinical research. One platform? Yes, and without a fork. Policy is evaluated per action rather than per application, so two regimes with different rules run on the same runtime with different checkpoints enforced. Maintaining two platforms to satisfy two regulators is the expensive way to solve this. ### Do you have an office in Houston? 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 one hour behind you from November to March and two hours behind from March to November. A morning standup on your clock is still early afternoon on ours. Most delivery is remote or hybrid. We travel for discovery, workshops and the opening weeks of a build.