CURRENTLY BUILDING PRODUCTION AI SYSTEMS

Hi! I'm Rob. An AI Engineer.

I'm Rob Dalida — an AI Engineer at Frontline Education. I own production AI systems end to end: data, model behavior, deployment, monitoring, and the cross-functional decisions that make them useful.

ROLE AI Engineer III
EMAIL robertdalida@gmail.com
FOCUS Agentic workflows
Rob Dalida portrait
EXPERIENCE
8yrs ops·2yrs AI/ML
SHIPPED
11prod systems
Python TypeScript Next.js FastAPI AWS Lambda AWS Bedrock AWS SageMaker Snowflake PostgreSQL D3.js ReactFlow OpenAI Vector Embeddings EventBridge Codex Claude Cognito CloudWatch GitHub Actions Terraform CI/CD Salesforce Zendesk Pendo AWS RAG
01 / ABOUT

The bridge.

domain × engineering

I spent years inside the messy reality of school-based Medicaid claiming — IEP documentation, compliance, client operations, the workflows that software is supposed to make easier. Now I build the AI systems that actually do.

02 / OUTCOMES

Numbers that moved.

measured, not claimed
0%
// initial-prompt tokens

Redesigned a broad agent into orchestrator + specialists. Same prompts, far less context.

0%
// answer coverage

Auto-research loop took an Absence Mgmt eval from ~23% to ~80% — bounded, reversible changes only.

0→170
// security findings

5 days with Snyk MCP + repo harness. All 10 critical findings cleared, triaged the rest.

0.000
// rag relevancy

Resolution Cards lifted mean relevancy from 0.912 → 0.976 across 30 support scenarios.

03 / ACTIVITY

Work you can't always see.

private repos · 2025–2026
2,391 contributions in the last year

Last synced: Apr 27, 2026

Private repository work included in this local snapshot.
04 / OPERATING SCOPE

Beyond the title.

architecture · delivery · adoption

Operating across architecture, delivery, and adoption. The work has grown into broader system ownership: defining direction, creating reusable AI patterns, and helping cross-functional teams turn hard ambiguity into production workflows.

// system owner

Data to monitoring.

Own the whole AI path: source data, model behavior, deployment, production checks, and iteration after launch.

// architecture driver

Direction, not just delivery.

Propose architecture, evaluate tradeoffs, and push decisions that make systems safer, reusable, and easier to operate.

// org multiplier

Patterns others can reuse.

Standardize evaluation metrics, repo harnesses, agent workflows, and pipeline practices so teams can move faster with less guesswork.

// cross-functional lead

Hard problems across teams.

Partner across Support, Product, Legal, GTM, and Services to translate business pressure into practical AI systems.

05 / SELECTED WORK

Things I've built.

production · 2024–2026

Apps I've built.

product surfaces · internal platforms · cross-functional tools

Automations I've shipped.

workflow removal · compliance · operations

AI Infrastructure I've standardized.

evals · agents · harnesses · reusable knowledge systems
Knowledge Graph

Content Graph

An AI knowledge graph builder that takes Learning Center articles, embeds them, and surfaces missing links to optimize content for AI retrieval and support deflection.

OpenAISemantic EmbeddingsD3.jsAirtableNext.js
Portfolio mockup
Visual Builder

Flow Builder

A ReactFlow drag-and-drop workflow builder for data transformation — supporting Financial Cost Report calculations and no-code-style processing pipelines.

ReactFlowTypeScriptModular Nodes
Portfolio mockup
UploadCSV/XLSX
</>LogicDATA_PARSER
TransformDATA_TRANSFORM
AggregateANALYZE_DATA
Compliance

PII Redactor

Production Zendesk SSN redaction with scanner/redactor phases, dry-run safety, backups, daily Lambda automation, and stakeholder reporting. Cleared an 88-ticket / 180-day backlog at launch.

AWS LambdaEventBridgeZendesk
Eval Quality

Resolution Cards

Structured knowledge extracted from solved tickets. Lifted mean relevancy 0.912 → 0.976, faithfulness 0.935 → 0.963, and eliminated relevancy failures from 16.7% → 0%.

Agentic

Auto-Research Loop

A governed loop that scores knowledge quality, proposes one bounded change, reruns deterministic eval, and keeps it only if quality improves — otherwise restores prior state.

23 runs 14 kept 80% best
Kept improvement Discarded Running best
CodexEval HarnessMarkdownSource-Grounded
Security

Snyk MCP

Connected Snyk MCP to Codex for security triage, remediation, verification, and rescans. In 5 days, dropped findings 468 → 170 and cleared all 10 criticals.

468170
5-DAY DELTA · 10 CRITICALS → 0
Agentic Harness

Repo Agentic Harness

A reusable repo contract for AI-assisted engineering: clear entrypoints, ownership boundaries, verification commands, secret stop conditions, and safe handoff rules.

CodexVerificationAGENTS.md
ML Platform

Model Workbench

A Machine Learning platform for expense classification — datasets, training/eval concepts, model registry patterns, and a playground for testing classification logic.

SageMakerS3PythonFastAPI
Portfolio mockup
Actual Costs Predicted
Confidence98.4%
Travel $2.8K Supplies $0.9K
Compliance Engine

Rules Engine

A nationwide compliance engine that sources state and federal regulatory documents, translates them into structured outputs, and helps product and engineering create business rules.

PythonAWS BedrockReact
Portfolio mockup
TexasIEP ProcessMedium ConfidenceApproved
IF IEP.includes(independentSupports) THEN
  IF serviceType == 'direct' THEN
    verify schedule AND access
  ELSE verify student-specific support
END IF
AI Observability

Signal Console

Real observability for AI support systems — anomalies, retrieval behavior, guardrails, and incidents in a single operational surface the team can open every morning.

Portfolio mockup
signal-console / overview

Overview

Monitor chatbot performance and conversation signals.

Engagement 276
Users 197
Positive 67%
SourcesInteractive
Popular Questions
Pre-Built Reports
Service Log
Pipeline Intelligence

Pipeline Scoring

A GTM and Sales pipeline intelligence platform that ranks accounts, identifies expansion opportunities, and highlights clients with upsell potential.

Portfolio mockup
ScenarioBaseline V1K-12 District
100Ranked Accounts
$43.9MQualified Pipeline
42Triggered Accounts
Scoring Workbench · Live Preview Unsaved
Fit 2.0×

Ideal customer profile match

Pathway 1.0×

Source → target solution path

Peer Reference 1.7×

Nearby customer references

Propensity 1.0×

Likelihood to buy signals

Trigger 2.0×

Recent trigger events

Risk Penalty 1.0×

Risk factor weighting

Legal Automation

COI Automation

An AWS-based Certificate of Insurance pipeline that replaced an annual manual legal workflow with secure storage, scheduled processing, audit trails, and personalized email + PDF delivery.

AWSS3EventBridgeSES
Ops Automation

OKR Prep

A browser-local workflow for cleaning, normalizing, validating, and previewing OKR import files before they move into downstream planning systems.

CSVValidationPreview
Learning Ops

Docebo User Provisioning

An automation path for preparing and provisioning learning-platform users, reducing manual account setup work and improving consistency across imports.

User ImportValidationLMS Ops
Knowledge Infrastructure

LLM Wiki

A markdown knowledge-base pattern where raw sources stay immutable, synthesized pages stay maintainable, and agents can navigate durable context without vector-first complexity.

MarkdownIndexingSource Grounding
Agent Platform

Mastra Agents

Agent workflows for routing tasks, preserving context, and turning repeated AI work into reliable, inspectable execution paths instead of one-off prompts.

AgentsWorkflowsObservability
Repo Knowledge

Repo Wiki

A Codex-native workflow that turned real internal repositories into browseable, refreshable documentation, including 62 markdown pages across 3 repos.

CodexDocsRepo Maps
06 / RECOGNITION

What partners say.

adoption · trust · impact
"Rob took the time to understand our workflows, listened thoughtfully, and delivered a practical, effective solution."
Sr. LMS Administrator, TL
"The team resolved hundreds of vulnerabilities within the first week, making an immediate reduction in risk exposure."
Sr App Security Engineer
"Rob turned a tedious quarterly process into a clean, reliable import workflow, saving hours every quarter and giving the team confidence in the data."
VP, Strategic Init
"Rob brought the vision to life quickly, beautifully, and with real partnership."
Director, GTM Strategy & Plan
"Rob brought a complex automation across the finish line with patience, care, and follow-through."
Sr. Paralegal
07 / FIELD NOTES

What I've learned.

editorial notes · experiments · reuse
Knowledge Systems

Creating a Self-Maintaining Knowledge System

A source-grounded markdown wiki pattern for keeping internal documentation structured, reviewable, and easier to improve over time.

CodexMarkdownKnowledge Ops
Read note →
AI Product Pattern

How Codex Skills Recreated a SaaS-Style AI Product

A repo-wiki workflow that turned real internal repositories into browseable, refreshable documentation without needing a separate AI SaaS tool.

CodexSkillsRepo Wiki
Read note →
Eval-Driven Improvement

AI Auto-Research Loop Improves Retrieval Quality

A governed loop that proposes one bounded knowledge-base change, scores it, and keeps only improvements that pass deterministic checks.

CodexEvalsRollback
Read note →
Support Quality

Resolution Cards Improve AI Support Response Quality

A test of operational knowledge from solved tickets, showing higher relevancy and faithfulness than documentation-only retrieval.

ChatGPTClaudeSupport Evals
Read note →
Agent Fundamentals

Anatomy of an AI Agent

A practical mental model for how LLMs become useful agent systems through memory, context, tools, prompts, planning, and evaluation.

LLMsRAGAgent Design
Read note →
Workspace Design

Harness Engineering: Structuring Workspaces for AI Agents

A repository design pattern that gives AI coding agents clear entrypoints, workflow contracts, verification paths, and stop conditions.

CodexGitHub CoPilotWorkflows
Read note →
Context Engineering

Context Engineering for Better AI Answers

A retrieval pattern for giving LLMs the right slices of context instead of overwhelming them with long, noisy prompts.

LLMsRAGRetrieval
Read note →
Agent Orchestration

Improve Token Usage with Agent Orchestration

A specialist-agent routing pattern that reduced token usage while preserving the feeling of one visible assistant experience.

CodexClaudeRouting
Read note →
08 / TRAJECTORY

The path here.

ops → ai engineer iii
2025 — NOW
FRONTLINE EDUCATION

AI Engineer III

Customer Experience AI · CX-AI
Own end-to-end AI systems from data through model behavior, deployment, monitoring, and iteration. Drive architecture decisions, standardize evaluation and tooling patterns, and partner across Support, Product, Legal, GTM, and Services on hard AI problems.
2022 — 2025
FRONTLINE EDUCATION

Medicaid Services Manager

School-based claiming · operations
Owned IEP documentation, claiming workflows, compliance, and client operations. Saw firsthand where software helps and where it gets in the way — that perspective is now my engineering compass.
2020 — 2022
FRONTLINE EDUCATION

Medicaid Services Team Lead

Medicaid services · team leadership
Led service operations across Medicaid workflows, supported team execution, and helped translate client needs into repeatable operational patterns.
2018 — 2020
FRONTLINE EDUCATION

Medicaid Client Manager

Medicaid services · client operations
Managed Medicaid client relationships and day-to-day operational support, building the domain foundation that later shaped my AI engineering work.