Document intake pipelines
Turn PDFs, emails, and images into typed, validated data.
Scoped Build · 60-90 days to production
Production AI for one bet that matters. Director-altitude scope, shipped to your team.
One AI use case, scoped against your Three-Horizon allocation, built to production in 60-90 days. Your team owns the eval suite, the runbook, and the on-call after handoff. Not an automation shop. Not a freelance build.
TL;DR · 60-second read
What you walk away with
Production AI in 90 days, defended by the eval suite your team owns.
- Who it's for
- Operators with one AI bet that matters and an engineering team ready to own it after handoff.
- Format
- 60 to 90 day build inside your stack. Eval suite, runbook, and on-call handed to your team.
- Investment
- Quoted in writing within 48 hours.
- Next step
- Book the Build call
What you get
One production AI capability, deployed inside your stack.
Owned by your team after I leave. Model-agnostic by default, structured around patterns I ship in Baxie today. If your use case isn't listed, ask. The patterns transfer.
Multi-agent estimating, scoping, or generation workflows
Match how Baxie's orchestrator works.
AI copilots inside existing tools
CRM, ticketing, intake forms that augment a team without replacing the workflow.
Closed-loop learning systems
Production output becomes training signal for the next run.
Retrieval-augmented assistants
Grounded in the shared knowledge base your team and your AI both work from, or built alongside one.
The standard
Production reliability is the deliverable, not the showcase.
Most "AI build" consultancies optimize for the demo. This engagement optimizes for the steady state.
Without · What most consultancies do
The demo handoff
Build the demo, hand off the code, charge for the deck. Skip eval. Production reliability is your problem. You're left with undocumented prompts and a brittle pipeline.
With · What you get here
The steady-state handoff
Production code in your repo, cloud, accounts. Eval suites that catch regressions before users do. Multi-model routing that controls cost without manual tuning. Observability per environment. Documentation written for the team that takes it over.
The engagement
60 to 90 days, four phases. Daily Slack, two working sessions per week.
Me as builder. One of your engineers rides along as future owner. One product or ops partner owns the use case.
Phase 1 · Weeks 1 to 2
Scope and ground truth
Lock the use case to one workflow. Collect 50 to 200 real examples for evals. Pick the model strategy against cost and latency targets.
Phase 2 · Weeks 3 to 6
Build the spine
Stand up the agent graph, routing, retrieval, observability. Wire production integrations to your tools. Eval suite running against real examples.
Phase 3 · Weeks 7 to 10
Production hardening
Token tracking, rate limit handling, retries, stall detection. Multi-model routing (Opus hard, Sonnet default, Haiku routine). Regression suite on the prompts that matter.
Phase 4 · Weeks 11 to 12
Handoff
Train the team on the eval workflow. Document the playbook. Schedule 30/60/90 check-ins.
The approach
Model-agnostic by default. The patterns transfer.
If you're on a different stack, I'll learn yours in the first week. The eval discipline, the routing logic, and the observability patterns travel.
Model and protocol layer
Anthropic Claude API (Opus, Sonnet, Haiku), OpenAI, Model Context Protocol (MCP).
Application layer
Python and TypeScript, Next.js Server Actions, Supabase, Postgres.
Retrieval layer
pgvector, vector databases, RAG pipelines.
Eval and testing
Braintrust for evals, observability, and prompt regression. Playwright for E2E coverage on AI-touching UI.
The outcome
Week 12, you ship. Six months later, the spine compounds.
The handoff is the engagement. By week 12, your team can debug and iterate without me.
Week 12 · What ships
One AI workflow, in production
Owned by your team. Eval coverage on the top 5 to 10 failure modes. Token tracking and observability per environment.
+6 months · What clients report
The spine compounds
Workflow stays in production without the original consultant on retainer. New AI features build on the spine instead of starting from zero. Cost stays controlled because the routing is documented.
The guarantee
The guarantee.
Ground truth before scope. Both parties sign the spec by day 14, or the engagement ends and you walk with the artifacts. That removes the fixed-bid blind-bet risk from the buyer's pile.
Scoped AI Build guarantee
Spec signed by day 14 or you walk
By end of day 14, both parties sign the spec sheet (use case, ground-truth dataset, eval baseline, integration plan, success criteria, model strategy). If not signed, engagement ends. Client keeps the eval scaffolding, the ground-truth dataset, and the written read on why it stalled. Days worked are billed at the day rate, not the build rate.
Umbrella · Every engagement
14-day mutual exit
Every engagement carries a 14-day mutual exit. Either party can end the engagement inside the first 14 days. Days worked are billed at the agreed day rate. No further commitment.
What's included
What's included beyond the build.
The build is the deliverable. These come with it.
Bonus 01
Production runbook and on-call documentation
The manual your next on-call engineer needs. Stall detection, retry logic, failure modes, escalation paths. Written for the team that takes it over.
Bonus 02
30 days of post-launch on-call from Edwin
After handoff, 30 days where production incidents come to me first. Bug triage, prompt regressions, model behavior shifts. Your team learns the runbook by watching it run.
Bonus 03
Team handoff training session, recorded
Live walkthrough of architecture, evals, routing, failure modes. Recorded so the next hire onboards from the tape.
Bonus 04
Eval suite the team owns and runs after handoff
Versioned eval set with labeled examples, regression tests on the prompts that matter, CI wiring on every prompt or model change. Quality bar lives in the repo, not in my head.
FAQ
Frequently asked questions
What is a Scoped AI Build?
A 60 to 90 day engagement that ships one production AI capability inside your stack: multi-agent workflow, document intake pipeline, AI copilot, retrieval-augmented assistant, or closed-loop learning system. Deliverable is production-deployed code in your repo, your cloud, your accounts, plus eval suites, multi-model routing, observability, and documentation. Your team owns the on-call after handoff. Not an automation shop. Not a freelance build.
How much does a Scoped AI Build cost?
60 to 90 days end to end. Includes a 14-day ground-truth phase: if the data doesn't support production reliability, you keep the eval scaffolding and we end the engagement. Quoted in writing within 48 hours of the discovery call.
How is a Scoped AI Build different from hiring an AI automation agency?
Automation agencies optimize for the demo. Scoped Build optimizes for the steady state. Production-deployed code in your repo (not theirs), eval suites that catch regressions before users do, observability per environment, multi-model routing that controls cost. The team owns the runbook after handoff, no undocumented prompts or brittle pipelines.
What kinds of AI use cases fit a Scoped Build?
Document intake pipelines (PDFs, emails, images to typed validated data), multi-agent estimating or scoping workflows, AI copilots inside existing CRM or ticketing tools, retrieval-augmented assistants grounded in a shared knowledge base, and closed-loop learning systems where production output becomes training signal. Model-agnostic by default. The patterns transfer to other use cases too.
We have an idea but no AI engineer in-house. Can you still do this?
Yes. The engagement assumes one engineer from your team rides along, not as the builder, but as the future owner. If you don't have anyone yet, we add a hire-or-train phase upfront.
What if the use case turns out to be a bad fit for AI?
You find out in week 2, not month 6. The first phase is ground truth. If the data doesn't support production reliability, I tell you, you keep the eval scaffolding, and we either reframe the use case or end the engagement.
Do you build with our existing AI vendor?
Yes. The build is model-agnostic by default. Bring your stack. The patterns transfer.
Is there a maintenance retainer after?
Optional. Most clients move into a fractional retainer if they want me to keep owning AI work post-build. Otherwise the team runs it.
What's your guarantee on this engagement?
Two guarantees. SKU guarantee: by end of day 14, both parties sign the spec sheet (use case, ground-truth dataset, eval baseline, integration plan, success criteria, model strategy). If not signed, engagement ends. Client keeps the eval scaffolding, the ground-truth dataset, and the written read on why it stalled. Days worked are billed at the day rate, not the build rate. Umbrella guarantee on every engagement: a 14-day mutual exit. Either party can end the engagement inside the first 14 days. Days worked are billed at the agreed day rate. No further commitment.
Pick the bet
Pick the bet. I'll ship it.
Book the call. 30 minutes to scope. Two weeks to start.