About
TH Intelligence builds practical AI and automation for healthcare organizations, helping medical offices turn their data into better clinical, operational and financial decisions.
Who's behind it
Thiago Bandeira
General Manager and Head of AI at a primary care practice · Miami, FL
I'm Thiago Bandeira, General Manager and Head of AI at a primary care practice.
I build the AI agents, machine-learning tools, RPA bots and data systems the practice runs on, and I use them every day alongside the people they're for.
The open-source projects are the same method in public: rules where it's audited, AI where it drafts, a person where it decides.
TH Intelligence was founded in 2026 so that other practices can get the same tools without building them from scratch. The chart-review bot, the coding-gap engine, the finance dashboard and the operations app on this site were not built for a client and handed over. They run inside a working practice, during clinic hours, with the people who use them a few desks away. What breaks gets fixed the same morning, and what gets measured gets written down.
RPA bot means robotic process automation: software that clicks through screens the way a person does. It is not AI; it is the right tool for clicking, reading and matching.
Background
- Education
- M.S. in Data Science and Artificial Intelligence, Florida International University
- Certifications
- AWS Certified Generative AI Developer (Professional); AWS Certified Machine Learning Engineer (Associate)
- Languages
- English, Portuguese, Spanish
What I build
Four kinds of work, each with the case study that shows it.
AI agents
Drafting, summarizing and sorting, with a person approving before anything reaches a patient or a payer.
Coding gaps, with the evidence attachedMachine-learning tools
Risk scores and worklists that show their reasons, built on the practice's own data.
One risk score anyone can read the reason forRPA bots
Software that clicks through the EHR and portals the way a person does, for the reading and matching work.
Chart prep without the clickingData systems
Importers, one view per member across health plans, and live dashboards for every department.
One view of every member, across three health plans
In public
- hedis-care-gap-agent
A rules engine finds care gaps, AI drafts the outreach, and nothing goes out until a person approves it.
- hedis-spec-copilot
Answers questions about public HEDIS-aligned measure guidance and cites the source passage for every factual sentence.
- clinical-agent-evals
A test harness for clinical AI agents where every number is measured, never typed in.
- fhir-feature-service
Turns FHIR patient records into model-ready features that can never see the future.
- HCC Coding Assistant
Clinical language in, validated ICD-10 codes and a CMS-HCC category out, with every number coming from official CMS files rather than the model's memory.
What I'm working on
- Agentic AI for clinical documentation and risk-adjustment review.
- Real-time quality-gap closure that plugs into care-team workflows.
- Machine learning that survives contact with production healthcare data: messy scans, sparse labels, and all.
How we work
Clear rules handle anything a payer or auditor will review. AI does the drafting, summarizing and sorting, and a person approves before anything reaches a patient or a payer. Every engagement starts by measuring how the work is done today, so results can be verified, not just claimed. Solutions are built to fit the systems a practice already uses and to protect patient data at every step.
Who we serve
Our focus is value-based care practices: primary care groups working with Medicare Advantage, ACO and health plan contracts. We also support specialist offices and hospitals that want to reduce manual work and get more out of the data they already have.
What this site will never do
- Name the practice, its staff, its patients, or the health plans it works with.
- Show a real system. Every screen on this site is a mock-up with sample data.
- Quote a number that isn't in a log or a README, or invent a client, a testimonial or a result.
- Claim HIPAA compliance, guaranteed savings or a return on investment.
- Call an RPA bot “AI”.
Start with 20 minutes.
Bring one process that eats your team’s day. We’ll talk through what to measure first, and whether automation or AI fits it at all.