Research

Carson Rodrigues

Independent researcher working at the intersection of production AI systems and their measurement — voice-AI latency, LLMOps, the Model Context Protocol, multi-agent reliability, human–AI trust, and clinical ML. Nine papers — one accepted at ICANN 2026 (Springer LNCS), with five more under peer review. Most studies are built on systems I ship in production, so the methods are evaluated under real workloads rather than in isolation.

iD0009-0001-7195-6742carson@celabe.comAffiliation: Celabe, Research Division
ORCIDGoogle ScholarOpenReviewOSF / PsyArXivResearch site
9
Papers
1
Accepted
5
Under review
2
Live preprints
Voice AI systems & latencyLLMOpsModel Context Protocol (MCP)Multi-agent reliabilityHuman–AI trustAgentic safetyClinical ML

Papers & preprints

Updated June 2026
AcceptedICANN 2026·Conference · Springer LNCS

Latency Optimization for a Production Voice AI Platform

Carson RodriguesiD (Celabe), Oysturn VasiD (University of Waterloo)

A systems-level latency study of a production voice-AI platform (Anthropic Claude intent detection + ElevenLabs TTS over a NestJS WebSocket pipeline). The central finding is that running intent detection and TTS concurrently — rather than shaving any single stage — is the highest-leverage optimization, cutting median end-to-end latency from 3,277 ms to 1,909 ms.

−41.8% p50 latency (3,277 → 1,909 ms)Voice AILatencyLLM systems

Accepted at ICANN 2026 (peer-reviewed; Springer LNCS proceedings). Registration confirmed; camera-ready June 2026.

Under reviewICSME 2026 — Industry Track·Conference

MCP Server Architecture Patterns

Carson RodriguesiD (Celabe), Oysturn VasiD (University of Waterloo)

A pattern catalogue for production Model Context Protocol (MCP) servers — a Gamma-format taxonomy, anti-patterns, and cross-cutting concerns — validated with a real inter-rater reliability study (κ = 0.76). Proposes a Proxy-Aggregator pattern once tool counts exceed a practical threshold.

Inter-rater κ = 0.76MCPSoftware architectureAgents

Resubmitted to ICSME Industry Track after a venue-mismatch reject; strengthened with the inter-rater study.

SubmittedEMNLP 2026 — Industry Track (Budapest)·Conference

When Do LLMs Replace Fine-Tuned NLU? A Decision Framework for Intent Detection

Carson RodriguesiD (Celabe), Oysturn VasiD (University of Waterloo)

A decision framework for choosing between LLM classifiers and fine-tuned NLU on noisy production transcripts. Shows that full-data TF-IDF still reaches 95.2% on ATIS, and maps the regimes where an LLM-based intent classifier is — and is not — worth its cost and latency.

95.2% on ATIS (full-data TF-IDF baseline)NLUIntent detectionLLM evaluation

Submitted via OpenReview; editable until 2026-06-17.

In revisionTMLR·Journal

LLM-OptFlow: LLM-Driven Hyperparameter Optimization

Carson RodriguesiD (Celabe), Oysturn VasiD (University of Waterloo)

An LLM-driven hyperparameter-optimization harness benchmarked head-to-head against Optuna-TPE and random search on PMLB datasets, with multi-seed runs and budget curves to isolate where LLM-guided search actually helps.

Optuna-TPE / random / LLM HPO benchmarkAutoMLHyperparameter optimizationLLMs

Rebuilding the harness after an initial reject; new multi-seed evaluation in progress.

SubmittedAI & Society (Springer)·Journal · review article · double-blind

Personality Traits and Trust in Large Language Models: A Scoping Review

Carson RodriguesiD (Celabe), Simran Marian Rebello (Celabe)

A scoping review of how personality traits shape user trust in large language models, extended toward agentic (action-taking) systems with a set of testable propositions for future empirical work.

Scoping review + agentic propositionsHuman–AI trustPersonalityScoping review
PsyArXiv preprintdoi:10.31234/osf.io/2mdu8

Submitted to AI & Society (Q1); currently in technical check. Preprint openly available.

SubmittedNeurIPS 2026·Conference · double-blind

Context Drift in Multi-Agent LLM Systems

Carson RodriguesiD (Celabe)

Introduces a Context-Drift Score (CDS) metric and a structured verification protocol (SSVP) for detecting and mitigating divergence in coordinated multi-agent LLM systems.

CDS metric + SSVP protocolMulti-agentReliabilityEvaluation

Under double-blind review; author notifications expected late September 2026.

Preprint liveClinical / dental-AI journal (in submission)·Journal

DentaCoPilot: Dental Procedure Prediction from Patient Records

Carson RodriguesiD (Celabe), Steffie Dione Rebello (KLE (co-PI))

A machine-learning approach to predicting dental procedures from patient records, with a public-benchmark evaluation and a planned retrospective validation on real clinical records (KLE), as larger real-world datasets are sourced.

Clinical-AI preprint · real-data validation in progressClinical MLHealthcare AIPrediction
medRxiv preprintdoi:10.64898/2026.05.07.26352635

Preprint live on medRxiv; real-data path now via KLE retrospective records + MEPS benchmark.

Under reviewIC-SIT 2026 (Silicon University, India)·Conference · IEEE Xplore

A Structural-Similarity (SSIM)-Based Framework

Carson RodriguesiD (Celabe), and collaborators (7 authors)

A structural-similarity (SSIM)-based framework developed as a seven-author collaboration, targeting IEEE Xplore conference proceedings.

Collaborative (Carson a co-author)SSIMCollaboration

Acceptance notification expected 2026-07-01.

Planning · preregistrationTop HCI venue (TBD)·Conference / journal

Personality and Over-Delegation to Agentic LLMs

Carson RodriguesiD (Celabe)

An empirical follow-up to the trust scoping review (Paper 05), testing its propositions on over-delegation to agentic, action-taking LLMs. Preregistration in progress.

Preregistered study (in design)Agentic safetyHuman–AI trustOver-delegation

Several papers are under double-blind review, so author lists and venues may be anonymized in the submitted copies. DOIs are linked where a preprint is publicly posted; full manuscripts, data, and code for any work in progress are available on request — carson@celabe.com. ORCID: 0009-0001-7195-6742.

Available for senior AI / contract / FDE work

Building something with AI?

Voice agents, MCP servers, LLM pipelines, agentic workflows — pick a slot, drop a message, or send your email and I'll reply within a day.

Book a 30-min callWhatsApp me
or leave your email

Replies within ~24 hours · Remote-first · global · open to relocation