TA-10080 REG: Preparing the Pacific Renewable Energy Investment Facility, Phase 3 - Senior Developer / Software Development Manager (E-049450-038)

1. Project objective

Objective and Purpose of the Assignment

Background and Rationale

This assignment covers two beta web applications developed by the ADB Energy Sector Office’s Emerging

Areas team: the Marine Renewable Atlas - Philippines (MRAP) and the Asia-Pacific Energy Data Repository

(APEDR-ARCS).

MRAP is an interactive, point-and-click pre-feasibility tool: for a selected site or administrative area

it (i) returns renewable-energy potential as a set of ranked options (e.g., wave, tidal-stream, floating

PV with battery storage), (ii) features a guided, non-generative in-tool assistant that helps users

identify candidate sites and set screening parameters, and (iii) can generate roadmap reports.

APEDR-ARCS is a companion open-data repository delivered as a map-first experience: an interactive Asia-

Pacific region / country map as the landing page, where clicking through regions and countries surfaces

geospatial layers on the map and visualizes tabular data over CKAN's retained catalogue search.

The two

platforms share one architecture: CKAN governs and catalogues the data, object storage (S3) holds the

bytes, and a lightweight tiler serves geospatial layers to the map at runtime.

The underlying data foundation for this assignment is already complete.

Data collection, harmonization,

validation, and integration have been completed using globally available datasets — including the

European Space Agency’s Copernicus Marine Service wave and ocean data, the World Bank's Global Solar

Atlas, TPXO9 tidal-stream harmonics— complemented by local layers sourced through the Philippines'

national geospatial data portal.

The resource model is operational and, with minor adjustments, is ready

to be deployed to a web application.

While the platforms are deliberately designed to be adaptive —

accommodating additional datasets as needs evolve and as the tool scales to the Pacific — this

engagement is a productization and platform-development effort, not a data-collection exercise.

Consultants are expected to build on the existing, validated data and model rather than to assemble or

re-derive the underlying datasets.

The engagement runs October 2026 – April 2027: the platform is built to an end-February 2027 beta

launch, followed by a dedicated QA/QC and refinement phase in March and April driven by user feedback

and updates to the underlying model.

Delivery is by a coordinated team of three individual consultants —

a Product Leader / CTO (150 working days), a Senior Developer / Software Development Manager (120

working days), and a Senior UI/UX Specialist (100 working days) — alongside collaboration with

University of Edinburgh on the underlying model.

Every consultant deliverable is reviewed and refined by

ADB for quality assurance before it is accepted.

AI functionality is guided and non-generative,

consistent with ADB's moratorium on public-facing generative AI.

Objective and purpose of the assignment:

The objective of this assignment is to build the MRAP service layer, deliver the application frontend,

the guided in-tool assistant, the menu-of-options output engine, integrate the economic model, lead the

precompute pipeline, and contribute to the shared geospatial data-engineering lane.

The core MRAP map

frontend sits with the Product Leader to balance the contracts;

this role integrates against it, working

under the Product Leader's API contract and merge authority.

Through March and April, the role refines

the platform against user feedback and swaps in the updated cost functions and thresholds from the

University of Edinburgh review.

Scope of Work

The consultant is responsible for the MRAP backend and application frontend, the guided assistant and

menu-of-options features, a share of the shared geospatial data-engineering work, and post-beta

refinement:

• MRAP application frontend (parameter panels, results and menu views), integrating against the core map

frontend built by the Product Leader.

• Lead responsibility for the precompute pipeline, COG generation and STAC catalogue.

• A share of the distributed geospatial data-engineering work (GDAL, rasterio, cloud-native formats).

• Implementation of the assess API and zonal statistics,

• The menu-of-options output engine: ranked technology options per site or administrative area and

integration of the economic model, including BESS sizing and cost for floating PV.

• The guided, non-generative in-tool assistant (site identification and parameter-setting) and roadmapreport generation, with indicative labelling and provenance.

• Post-beta refinement (March–April): user-feedback-driven changes and integration of updated model

thresholds and cost functions through the request-time configuration.

Reference technology stack:

The following baseline is established in the MRAP/APEDR concept work.

Well-justified alternatives may be

proposed subject to ADB approval, provided they remain open-source-first and integrate cleanly with the

shared architecture.

Detailed Tasks and/or Expected Output

(October 2026 – April 2027;

beta launch end-February 2027, followed by QA/QC and refinement in March–

April;

inclusive of an approximately two-week reduction over the December holidays)

B1- Geospatial data engineering (shared): Precompute pipeline, COG generation from model outputs, STAC

catalogue, and the precomputed zonal-aggregate batch to PSGC region/province/municipality.

Requires

demonstrated GDAL, rasterio and cloud-native geospatial experience.

B2- MRAP application frontend: Parameter panels, results and menu views, comparison charts and

application state — integrating against the core MRAP map frontend built by the Product Leader.

B3- MRAP Features: Guided in-tool assistant, menu-of-options engine & asynchronous reporting.

A guided,

non-generative in-tool assistant that helps users identify candidate sites and set screening parameters

(technology, thresholds, exclusions);

the ranked menu-of-options output engine;

and asynchronous report

generation (queue/worker PDF generation and signed-URL delivery), with edge-case handling, performance

hardening and support to the end-February beta launch.

No public-facing generative AI, per ADB's

moratorium.

B4- Assess API & economic-model integration: The /assess endpoint (point, polygon, admin-zone

selection), windowed zonal statistics against COGs, and integration of the full economic model (energy

yield, CAPEX, OPEX, LCOE with uncertainty bounds and comparators).

The floating-PV pathway includes BESS

sizing and cost to address intermittency and firming.

Implements the University of Edinburgh costfunction specification when applicable.

B5- Post-beta refinement & model updates: Implements user-feedback-driven changes after the beta, swaps

in the updated cost functions and thresholds from the University of Edinburgh review through the requesttime configuration, retunes the menu-of-options and BESS logic as the model matures, and completes

regression testing and documentation to end-April.

Reporting, coordination and quality assurance:

The consultant will report to 3 Energy Specialists and Director, Energy Sector Office (Emerging Areas).

The consultant will work to the API contract owned by the Product Leader;

all code merges are subject to

the Product Leader and ADB Administrator's review authority.

The consultant will coordinate with the Senior UI/UX Specialist on the assistant UX and menu-of-options

design, and with the University of Edinburgh on the cost-function specification integrated during March–

April 2027.

Ownership, intellectual property and open-source release:

• All engagements are administered by ADB.

ADB owns all outputs — code, deliverables, data and derived

assets;

the consultant receives access rights, not ownership of the outputs.

• Intellectual property is assigned to ADB, and code is additionally released under a named open-source

licence (MIT or Apache 2.0 for code;

CC BY 4.0 for documentation and derived data).

• All work is developed in repositories accessible to ADB at highest privileges from day one

• Infrastructure is defined as code and is a named deliverable with acceptance criteria, enabling

redeployment into an ADB-designated account.

• AI features in MRAP are guided, non-generative product features;

there is no public-facing generative

AI, consistent with ADB's moratorium and AI-use policy, and any assistant output is clearly labelled as

indicative and kept human-in-the-loop.

• Placeholder cost functions are implemented ahead of the University of Edinburgh cost specification so

the architecture is not blocked;

the numbers are swapped in during the March–April refinement phase as

the model matures.

Minimum Qualification Requirements

Essential

• Advanced university degree (Master's or equivalent) in computer science, software engineering,

geoinformatics or a related field;

a first degree with additional qualifying experience may be

considered.

• At least 8 years of professional experience as a senior full-stack / backend engineer, with

demonstrated production delivery in Python and API development (FastAPI or comparable) and a JavaScript

map frontend (MapLibre or Leaflet).

• Demonstrated GDAL, rasterio and cloud-native geospatial (COG/STAC) experience — a requirement for this

engagement.

• Experience integrating scientific or economic models (energy yield, cost/LCOE, uncertainty) into a

production service.

• Experience with asynchronous processing (queues/workers) and performance profiling.

Desirable

• Experience building guided, rule-based decision-support or parameter-assistant features (nongenerative) is preferred

• Storage/firming (BESS) modelling or energy-systems experience is preferred.

• Familiarity with CKAN, STAC tooling and the ENERGYDATA.INFO ecosystem is preferred.

Minimum General Experience 10

Years

Minimum Specific Experience (relevant to assignment) 8

Years

Regional/Country Experience Required

Desired

Not Required

Deliverables

Details Name Type Estimated Submission Date

Geospatial data engineering (shared) Others 30-Nov-2026

MRAP application frontend Others 31-Dec-2026

MRAP Features Others 31-Jan-2027

Assess API & economic-model integration Others 28-Feb-2027

Post-beta refinement & model updates Others 30-Apr-2027

Schedule and Places of Assignment (chronological and inclusive of travel)

Schedule Type Continuous

Intermittent

Max.

Working Days/Week Home Office 5

Field 6

City and Country Estimated Start Date Estimated End Date Other Details (use if place selected is Others)

Home Office, Other Search and Select: Place

01-Oct-2026 30-Apr-2027

Export to PDF

Export to PDF

Copyright (c) 1998, 2023, Oracle and/or its affiliates.

All rights reserved.

Copyright (c) 1998, 2023, Oracle and/or its affiliates.

All rights reserved.

Pick a Date

Loading...

Loading...

Extraction covers only the accessible source. Linked TORs and later addenda are not verified unless their text has been retrieved; a missing field never means no requirement.