//ARIEL MAGALSO · AI ENGINEER
It never guesses
Verdict is a live AI lead-qualification system: identity resolution, cited research, an evidence gate, deterministic scoring, and human approval — designed and built end to end by one person.
No signup. Seeded fictional data. Nothing is sent.
//BUILT WITH
//ABOUT THIS PROJECT
At its core: AI that knows when not to act
The numbers below are measured from the live pipeline and its evaluation suite — each one represents a failure mode this system was built to refuse.
Labeled eval cases
Development and held-out sets
Scores on thin evidence
The gate refuses to guess
Responsible outcomes
Never a fifth guess
Cost per processed lead
Haiku + cached scenarios
//THE FOUR OUTCOMES
One inbox. Four responsible outcomes.
Sufficient evidence, meets the ICP — assigned, drafted, ready for a rep.
- Evidence-cited score
- Territory routing
- Outreach draft for approval
Can't be scored responsibly — returns the exact questions that would unblock it.
- No number emitted
- Unblocking questions
- SDR verification task
Scored, doesn't meet the ICP — reason recorded, no rep time spent.
- Explicit rule reason
- Ruleset version recorded
- Zero outreach sent
Matches an existing record — activity attached, never a silent duplicate.
- Identity-first matching
- Activity attached
- No duplicate records
//THE THESIS
Every automation starts with the ways AI fails
An LLM asked to score leads scores everything a confident 78. Asked to research a company it doesn't know, it invents a plausible industry. Both write straight into the CRM, where fabrication becomes indistinguishable from truth. Verdict is designed backwards from those failures — every safeguard exists because the naive version visibly breaks without it.
GUIDED SCENARIO
The ambiguous lead
“Fieldwork Group” resolves to three candidate companies. Verdict declines to guess.
0
Scores emitted
3
Unblocking questions
1
SDR task created
//THE PIPELINE
Judgment work, staged and gated
//01
Resolve & validate
Normalize the submission, resolve contact and company identity, and check for duplicates before any model spend.
//02
Research with citations
Enrich company facts against approved sources. Every fact carries a source link and a verified / uncertain / conflicting status.
//03
Gate, then score
An evidence-sufficiency gate runs first. Past the gate, deterministic ICP rules produce an explainable score — the model never picks the number.
//04
Approve & write
Outreach drafts and CRM changes are proposed as diffs. A human approves; writes are idempotent and fully audited.
//THE DIFFERENCE
Naive AI qualification vs Verdict
LLM straight into the CRM
Verdict.
//THE MIND BEHIND THE WORK
Ariel Magalso
AI Engineer · Philippines
I independently designed and implemented Verdict's pipeline architecture, application, evaluation suite, database, operational monitoring, interface, and deployment. I turn ambiguous business workflows into reliable, human-in-the-loop systems where traceability matters as much as speed.
- AI workflow architecture
- CRM and operations automation
- Structured LLM extraction
- Evaluation and reliability testing
- Human approval workflows
- Monitoring, cost controls, and audit trails
//HIRING?
Looking for someone who ships AI you can trust
Available for AI automation roles and consulting — workflow design, LLM evaluation, and production automation with real safeguards.
//PROOF LOG
Milestones you can verify
//THE PORTFOLIO STATEMENT
“The AI meant to clean your pipeline shouldn't become the fastest way to poison your system of record.”
THE FAILURE THESIS VERDICT IS BUILT AGAINST
//WHY IT HOLDS UP
Four properties, enforced everywhere
Evidence-linked
Every company fact and score criterion cites a source a human can open and check.
Deterministic where it counts
The model extracts and classifies; plain code applies the rules and arithmetic.
Human-in-the-loop
Consequential actions — outreach, CRM writes, merges — wait for explicit approval.
Operable
Latency, cost, failures, and audit events are visible on a live operations page.
//ILLUSTRATIVE CALCULATOR
Measure the tradeoff, not the hype.
Adjust the assumptions for your own operation. These are not measured customer results — this demonstrates the calculation, seeded with the fictional example from the product plan (800 leads/month, 12 minutes of manual work per lead, 65% auto-eligible).
LEADS ELIGIBLE FOR AUTOMATIC HANDLING
per month
STAFF HOURS POTENTIALLY RETURNED
per month
STILL ROUTED TO HUMAN REVIEW
per month, by design — not a shortfall
ESTIMATED NET MONTHLY SAVINGS
$3,640 handling-cost reduction − $8 automation cost
Is this a real working system or a mockup?
A live pipeline. The demo runs real identity resolution, research, gating, and scoring against seeded fictional data. Guided scenarios are idempotent on their submission id, so repeat clicks return the already-completed result instantly and for free.
Why does it sometimes refuse to give a score?
A number produced on thin evidence is noise that erodes rep trust. If fewer than the configured floor of ICP criteria can be resolved with evidence, Verdict emits no score and instead returns the specific questions that would unblock scoring.
What stops the AI from inventing company facts?
Facts only enter the system with a source URL and verification status, and outreach drafts are checked against those sources. Unsupported claims are stripped before a human ever approves a message.
What happens if a researched page tries to manipulate the AI?
Page content is treated as untrusted data. One guided scenario plants a prompt-injection payload in a seeded source — the instruction is ignored, flagged in the audit log, and never reaches a draft.
Can I see the code and the evaluation results?
Yes — the source is on GitHub, and the evals page publishes the full scorecard, including both failure directions: false scores and false refusals.
//AI SAFEGUARDS SERIES
Three systems, one argument
Each one automates judgment work an LLM will confidently get wrong — and each is built against a bigger consequence than the last. The safeguards get harder as the cost of being wrong goes up.
-
//01
Provenance
Customer support
Recoverable 1 / 3- If it's wrong
- A customer acts on a policy that doesn't exist.
- So it can't
- Answers below the groundedness threshold are discarded, not softened. Refusal is a successful outcome.
Visit site
-
//02 You are here
Verdict
Revenue operations
Persists 2 / 3- If it's wrong
- A fabricated fact enters the CRM and becomes indistinguishable from truth.
- So it can't
- The model classifies; plain code does the arithmetic. Below the evidence floor, no score exists.
95% accuracy across 60 cases
-
//03
LedgerGuard
Finance operations
Irreversible 3 / 3- If it's wrong
- An unverified invoice reaches a payment run.
- So it can't
- The workflow holds no payment authority. People approve; bank changes verify out of band.
Visit site