CTO Q1 2026 - 12 Week Year Plan

"The train can only move as fast as the tracks it's built on."

12-Week Vision

No more blind spots. No more silence. No more noise. Just signal.

Q1 Constraints & Milestones

Feature flag switch (Week 1) - new provider live
Force app upgrade (Week 4) - Rollout fully deprecated
AlloyDB deadline: Mar 15 (Week 10)
Baby arriving: ~Mar 20 (Week 11)
Solo + Claude (support as available)

WSJF Prioritization Framework

Weighted Shortest Job First (WSJF) is a prioritization model from SAFe that calculates the economic value of doing a job sooner rather than later. It helps sequence work to maximize value delivery given limited capacity. Higher WSJF = do it first.

(Business Value + Time Criticality + Risk Reduction) ÷ Job Size = WSJF

BV Business Value: Revenue impact, user value, strategic alignment

TC Time Criticality: Deadline pressure, decay of value over time, first-mover advantage

RR Risk Reduction: Reduces technical/business risk, enables other work, removes uncertainty

Size Job Size: Effort in ideal days/weeks. Smaller jobs get priority bump

Scoring Scale: 1-10 (Fibonacci-ish: 1, 2, 3, 5, 8, 10)

WSJF ≥ 5.0 Do first WSJF 3.5-5.0 Solid priority WSJF < 3.5 Defer/stretch

Goal 1

Single Pane of Glass

Apache Superset dashboard with system health, business metrics, and predictive indicators - MCP integrated

1.

Deploy Apache Superset infrastructure (dev/staging)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 10 Foundation for all visibility TC: 5 Needed before data connections RR: 10 Unblocks entire goal

÷

Size: 5 Docker/K8s setup, auth, networking

=

WSJF: 5.0
2.

Connect data sources: production DB, infrastructure metrics, business KPIs
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Enables actual insights TC: 5 Mid-quarter timing RR: 8 Can't build dashboards without data

÷

Size: 5 Multiple connectors, credentials, testing

=

WSJF: 4.2
3.

Build core dashboards: System Health, Business Metrics, User Analytics
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 User-facing value TC: 3 Can slip if needed RR: 8 Enables proactive monitoring

÷

Size: 5 3 distinct dashboards, SQL/viz work

=

WSJF: 3.8
4.

Build MCP server for Superset (Claude can query dashboards/data)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Claude answers "how's the system?" TC: 3 Nice-to-have RR: 5 Completes Goal 2 vision

÷

Size: 3 API wrapper, auth, limited scope

=

WSJF: 5.3
5.

Configure alerting rules for anomaly detection
1 2 3 4 5 6 7 8 9 10 11 12

BV: 5 Proactive vs reactive TC: 3 Polish, not critical RR: 8 "No more silence" vision

÷

Size: 3 Threshold config, notification setup

=

WSJF: 5.3

Goal 2

Claude as Company Brain

MCP integrations complete - Claude answers questions about system and business data

1.

Connect Sentry MCP for error/exception queries
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Instant error visibility TC: 10 Flag switch Week 1 = NOW RR: 10 Catch issues before users

÷

Size: 2 Pre-built connector, config only

=

WSJF: 14.0 🔥
2.

Deploy custom MCP server: production data access (read-only)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 10 Claude answers data questions TC: 8 Core capability, needed early RR: 8 Foundation for all queries

÷

Size: 3 Schema design, read-only endpoints, auth

=

WSJF: 8.7
3.

Deploy custom MCP server: code/business logic queries
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Claude explains "why does X happen?" TC: 8 Valuable but not blocking RR: 8 Deep system understanding

÷

Size: 3 Code indexing, semantic search setup

=

WSJF: 8.0
4.

Document MCP architecture + create internal usage guide
1 2 3 4 5 6 7 8 9 10 11 12

BV: 5 Team enablement TC: 2 Can be done anytime RR: 2 System works without docs

÷

Size: 3 Architecture diagrams, usage examples

=

WSJF: 3.0

Goal 3

Platform Play - Stripe for Labor

External developers can integrate with GigSmart via documented GraphQL API, OAuth, and MCP - FIRST TO MARKET

1.

Design + implement OAuth 2.0 flow for third-party authentication
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Security gate for all external TC: 10 🔑 Blocks MCP - critical path RR: 8 No shortcuts on auth

÷

Size: 5 Token flow, scopes, refresh logic

=

WSJF: 5.2
2.

Build external MCP server (GigSmart as MCP provider)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 First staffing MCP = moat TC: 8 🏁 Race to market - needs OAuth first RR: 5 Validates platform architecture

÷

Size: 5 Public server, rate limiting, docs

=

WSJF: 4.2
3.

Audit existing GraphQL schema - identify public-ready queries/mutations
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Defines what's exposable TC: 3 Independent track, flexible timing RR: 5 Security review prevents rework

÷

Size: 2 Schema review, security checklist

=

WSJF: 8.0
4.

Generate GraphQL documentation (schema docs + developer portal draft)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 5 Developer enablement TC: 3 Needed when partners onboard RR: 3 API works without docs initially

÷

Size: 3 Auto-gen + examples + portal setup

=

WSJF: 3.7
5.

Create developer sandbox environment for GraphQL testing
1 2 3 4 5 6 7 8 9 10 11 12

BV: 5 Safe testing environment TC: 3 Partner-dependent timing RR: 3 Isolation prevents prod issues

÷

Size: 3 Test data, isolated env, reset scripts

=

WSJF: 3.7

Goal 4

Smart Hire v2

Predictive worker no-show model using Axon + Claude-driven automatic experimentation

1.

Define training data requirements + extract historical shift/worker data
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Data quality = model quality TC: 5 Needed before training RR: 5 Unblocks all ML work

÷

Size: 3 SQL queries, data validation, feature design

=

WSJF: 6.0
2.

Set up Axon ML pipeline infrastructure
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Foundation for native Elixir ML TC: 3 Learning curve, needs time RR: 5 Enables training runs

÷

Size: 5 Nx setup, data loaders, training scripts

=

WSJF: 3.2
3.

Build Claude experimentation loop (auto-suggest/run model iterations)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 10 10x iteration speed, novel approach TC: 2 Differentiator, not urgent RR: 8 Autonomous improvement

÷

Size: 5 Prompts, eval harness, automation

=

WSJF: 4.0
4.

Train initial no-show prediction model (baseline)
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Direct business impact TC: 3 Needs data + pipeline first RR: 8 Proves concept works

÷

Size: 5 Model tuning, validation, iteration

=

WSJF: 3.8
5.

Deploy model to staging + create evaluation dashboard
1 2 3 4 5 6 7 8 9 10 11 12

BV: 8 Real-world validation TC: 3 Q2 production target RR: 5 Enables A/B testing

÷

Size: 5 Serving infra, monitoring, dashboard

=

WSJF: 3.2

12 Week Overview

Jan 6

Jan 13

Jan 20

Jan 27

Feb 3

Feb 10

Feb 17

Feb 24

Mar 3

Mar 10

Mar 17

Mar 24

Flag Switch

Force Upgrade

AlloyDB

Baby

CTO 12 WEEK YEAR PLAN

Q1 Constraints & Milestones

WSJF Prioritization Framework

12 Week Overview