Izzicupo Borges

Praxis

The operating system for what can't fail. Six products on one live model of your technical organization.

Talk to a founder

Deployed by our practice, alongside your team. Not sold on its own. It stays when we leave.

praxis / your-organization / chief-of-staff
Live

Today

  • 08:02

    Morning digest sent to the CTO

    3 priorities · 1 decision needed today

  • 08:14

    Cost anomaly · image-service

    +23% since Tuesday · owner notified

  • 09:40

    Review queue · 3 PRs waiting 48h+

    Thursday release at risk · reviewers nudged

  • 10:05

    Delivery · scheduling module

    6 days behind commitment · replan proposed

  • 11:30

    Readiness · SOC 2

    3 evidences missing · tickets opened with owners

  • 16:00

    Weekly update for the CEO

    Drafted · awaiting approval

Reviewing 214 open items across 9 connected tools…

Signals · last 30 days

Delivery flow

vs. last month

+18%

Cloud spend

−9% month over month

$41.2k

Review latency

+12% · above target

31h

Change failure

−2.0 pt

4.1%

Readiness

SOC 2
91%
HIPAA
100%
LGPD
86%

Architecture

Everything reads from the same model. Nothing to migrate.

Operated by

Your team, with our practice embedded

Praxis Graph

The live model of your technical organization

GitHubGitLabJiraLinearClickUpSlackTeamsAWSGCPAzureDatadogSentryKetryxVantaAny MCP server

Connected through MCP, with native ingestion only where continuous history needs it.

/01

Praxis Graph

The live model of your technical organization.

Everything your technical function uses connects here, through MCP. People, teams, services, deploys, costs, decisions, incidents and the requirements they serve become one model that every other product reads from.

  • Connects any tool with an MCP server
  • Ownership and dependency map
  • Decision and incident memory
graph / serviceTier 1 · clinical

scheduling-api

Service · 41 relationships

Owner
Julia Costa · Staff engineer
Team
Patient Experience
Clinical flow
Appointment scheduling
Monthly cost
$3.1k · −4%
Incidents · 30d
2 · none patient-facing
Requirements
14 traced · 2 without tests
Last deploy
2h ago · Praxis Build #493
Depends on
auth-service, notifications, ehr-gateway

/02

Praxis Chief of Staff

An agent that goes to the person.

Works like a chief of staff for the whole department: morning digests for leadership, nudges to the right engineer at the right moment, follow-ups on commitments made in meetings, and the weekly update for the CEO.

  • Daily digests and weekly updates
  • Proactive nudges in Slack or Teams
  • Commitments tracked from meetings
# eng-platformSlack
P

PraxisAgent09:40

@daniel — 3 PRs have been waiting on your review for 48h+. #482 blocks Thursday's release.

Open #482Snooze 2h
D

Daniel09:52

On it.

P

PraxisAgent10:31

#482 merged. Thursday's release is back on track. CTO digest updated.

/03

Praxis Signal

Delivery, quality and team health, measured continuously.

Flow, cycle time, review latency, change failure and team load, measured from the work itself rather than from spreadsheets, and tied to the business outcomes agreed for the engagement.

  • Delivery flow and predictability
  • Quality and incident trends
  • Team load and burnout risk
signal / deliveryLast 8 weeks

Work shipped per week

Cycle time

3.2d

−18%

Review latency

31h

+12%

Change failure

4.1%

−2.0 pt

/04

Praxis Cloud

Every cloud account and cost center in one view.

Connect AWS, GCP or Azure and see spend by service, team, environment and feature, compared month over month, with anomalies caught the day they start.

  • Cost by service, team and feature
  • Anomaly detection
  • Rightsizing recommendations
cloud / spend by serviceOctober

Total

$41.2k

−9% month over month

image-service

+23% since Tue

$12.4k

core-api

stable

$8.1k

data-pipeline

stable

$6.7k

ehr-gateway

−6%

$5.2k

staging

62% idle · rightsize

$4.9k

other

14 services

$3.9k
AWS · productionAWS · stagingGCP · analytics

/05

Praxis Readiness

Audit and regulatory readiness as a daily signal.

HIPAA, LGPD, SOC 2, PCI DSS, FDA and ANVISA readiness tracked continuously, with evidence pulled from the graph and from the compliance tools you already use, such as Ketryx and Vanta.

  • Continuous evidence collection
  • Gaps opened as tickets with owners
  • Integrates Ketryx and Vanta
readiness / frameworksAudit in 15 days
SOC 2
91%
HIPAA
100%
LGPD
86%
  • ✓Access review · October
  • ✕Change approvals · 2 missingticket opened
  • ✓Incident response drill
  • ✓Sensitive-data scan · logs and tickets
  • ✕Vendor assessment · 1 pendingticket opened

/06

Praxis Build

Engineering agents that ship through pull requests.

Agents that take well-scoped work — dependency upgrades, tests, migrations, small fixes — and deliver it as pull requests with checks. Every pull request gets its own ephemeral environment to review, which destroys itself when the work is merged.

  • Works from Jira, Linear or ClickUp
  • A preview environment for every pull request
  • Every change reviewed and approved by a human
build / pull requestsThis week
  • #491

    Upgrade openssl to 3.0.15

    Checks passing · preview environment ready

    Review
  • #493

    Add retries to the webhook consumer

    Approved by J. Costa

    Merged
  • #495

    Backfill audit fields on encounters

    Tests running

    Draft
  • #497

    Remove unused staging buckets

    Saves $1.1k/month · awaiting approval

    Review
Every change reviewed and approved by a human

Built for regulated work

The speed of software. The rigor of medicine, banking and aviation.

Teams in health, finance and aviation live between two forces that usually fight: the speed their market demands and the evidence their regulators require. Praxis makes the evidence a by-product of the work.

Readiness every day

HIPAA, LGPD, SOC 2, PCI DSS, FDA and ANVISA readiness tracked as a daily signal, instead of a three-month project before every audit.

Business context in the graph

Services and incidents are linked to the flow they affect — a clinical flow, a payment, a harvest, a flight — so priority follows real risk, not ticket size.

Agents designed for sensitive data

Agents are designed to detect health records, card data and personal information in logs, tickets and messages, and to keep them inside your boundary.

Playbooks from our own engagements

Regulated launches, audit preparation, HL7 and FHIR interoperability, clinical safety reviews and high-volume transaction processing, encoded from our own work.

Deployment

Praxis arrives with the engineers who run it.

It isn't sold off the shelf. A senior team from our practice connects it to your tools, runs it alongside your engineers and answers for the results.

01

Discovery

Praxis connects to your tools in read-only mode. Within days, you see the first real diagnosis of your technical organization.

02

Embed

Our engineers join your team and the agents start working alongside them.

03

Operate

The outcomes agreed up front are measured continuously in Praxis Signal.

04

Hand-off

Our team steps back. Praxis stays with you, under an annual license.

How our practice operates →

See Praxis on your own organization.

It starts with a conversation with one of our founders, and a discovery that runs on your real data.

Talk to a founder