Skip to content
Emmanuel Alabi
Menu
GitHubLinkedInContact

Emmanuel Alabi

Software Engineer & AI/Robotics Researcher

I'm interested in intelligent systems that can perceive, reason, learn, and act in complex environments.

My current focus is building deeper foundations in machine learning, robotics, computer vision, autonomous systems, and embodied intelligence.

Embodied AI · Robot Learning · Computer Vision · Autonomous Systems · Multimodal Intelligence

Research directionDownload CVGitHub

Current direction

Where the work is going, and how far along it actually is. Status labels describe maturity honestly; exploratory work is labelled as such.

View all
CompletedAugust 2026

Better Calibration Without Demonstrated Safety Gains: Risk-Calibrated Semantic Navigation Under Distribution Shift

Question. Does converting calibrated semantic perception uncertainty into navigation cost improve closed-loop robot outcomes under controlled distribution shift?

A preregistered ROS 2 and Gazebo benchmark testing whether converting calibrated semantic uncertainty into navigation cost makes a Nav2 robot safer under controlled distribution shift. Calibration improved; navigation safety did not.

Calibration improved substantially and clean-route efficiency was preserved, but the preregistered collision-reduction and severity-interaction hypotheses did not pass. Better component calibration did not establish safer navigation.

  • Embodied AI
  • Autonomous Systems
  • Computer Vision
  • Uncertainty Calibration
CompletedAugust 2026–September 2026

Early Failure Prediction and Recovery for Mobile Robot Navigation

Question. How early can mobile robot navigation failure be predicted from onboard signals, and can a guarded recovery act on that prediction without degrading outcomes?

A preregistered study of whether navigation failure can be predicted early enough to act on. Failures are detectable roughly four seconds ahead, but the learned predictor did not beat a transparent baseline at a fixed false-alert budget, and prediction-triggered recovery did not improve mission completion.

Both confirmatory hypotheses were not supported. Lead time and calibration held; the comparison against the transparent baseline, generalisation to unseen fault families, and the recovery benefit did not.

  • Embodied AI
  • Autonomous Systems
  • Failure Prediction
  • Uncertainty Calibration
CompletedAugust 2026–September 2026

Language as an Uncertain Sensor for Robot Navigation

Question. If a navigation instruction is treated as an uncertain observation rather than a command, can a robot recover from instructions that are wrong, ambiguous, or contradicted by what it sees?

A preregistered study treating a natural-language instruction as a noisy sensor rather than a command. It closed under a deliberately narrowed scope: a frozen upstream threshold made the preregistered calibration mathematically unreachable, and clean-view grounding saturated at zero errors, which removed the negative outcomes the planned models needed.

The preregistered four-class runtime calibration was not achieved and is not claimed. Two findings stand in its place: the pinned OCR stage truncates score support to [0.5, 1.0], so the required low-confidence bin cannot be populated at all; and human review of all 287 scoreable detections found every error confined to one class, leaving three classes with no negatives to fit.

  • Embodied AI
  • Multimodal Intelligence
  • Autonomous Systems
  • Uncertainty Calibration

Areas I'm exploring

These are interests, not claims of expertise.

More about me
Embodied AI
How intelligent agents can perceive, reason about, and interact with physical environments.
Robot Learning
Learning policies and representations for perception, navigation, manipulation, and control.
Computer Vision
Visual perception and representation for autonomous systems.
Autonomous Systems
Systems capable of sensing, planning, decision-making, and action.
Multimodal Intelligence
Combining vision, language, sensor observations, and action.

Selected projects

Implementations, engineering systems, and experimental work, distinct from formal research.

View all
Research EngineeringApril 2026–May 2026

Meta-Harness Plus: Experimental Search Over LLM Agent Harnesses

A framework for searching over LLM harnesses under a fixed compute budget: Pareto search across accuracy, tokens and latency; successive-halving evaluation; and drop-one ablation to attribute gains to individual components.

Python · Pareto optimisation · Successive halving · Ablation attribution

Academic ProjectJuly 2022–June 2023

BUCODEL: Learning Management System

A web-based learning management system for Babcock University's Centre for Open and Distance e-Learning, supporting student, lecturer, and administrator workflows across courses, assignments, materials, and class interaction.

React · JavaScript · Bootstrap · Node.js · Express · MongoDB · Mongoose · Socket.IO · Cloudinary

Systems

Obelo: Multimodal AI Infrastructure

A production AI platform combining multimodal generation, retrieval-augmented context, asynchronous orchestration, model routing, and evaluation systems.

Multimodal pipelines · Model routing · Retrieval-augmented context · Asynchronous orchestration · Evaluation systems

Engineering background

Before focusing more deeply on AI and robotics research, I worked across full-stack systems, AI infrastructure, distributed systems, developer tooling, and production software. That background is the reason I approach research as a systems problem as much as a modelling one.

  • Obelo — AI Systems & Multimodal Content Infrastructure

    Founder & CEO · Oct 2025 — present

    Designed and developed a production AI platform combining multimodal generation, retrieval-augmented context, asynchronous orchestration, model routing, and evaluation systems.

  • Punch Group — Production Web Systems

    Software Engineer · Jan 2025 — Jul 2025

    Delivered three production products: GraphQL query optimisation and memory-leak resolution on a Next.js/Express application, a Nuxt.js lead-generation tool with HubSpot and Salesforce integrations, and a real-time support service with in-app video.

  • SPay Business — Mobile & Financial Systems Engineering

    Software Engineer · Nov 2023 — Dec 2024

    Worked on production mobile-financial systems, including authentication improvements, application performance, interface redesign, CI/CD automation, and over-the-air mobile deployments.

  • Acadu — Educational Platform

    Founder · Feb 2022 — Sep 2025

    Led product, architecture and engineering: React Native mobile, Next.js web, NestJS backend, PostgreSQL, deployed on Fly.io and GCP with Redis and RabbitMQ for caching, message queuing and real-time event processing.

Full engineering history is on the CV