About Funga

Funga is a public benefit corporation harnessing forest fungal networks to address the climate crisis. We combine modern DNA sequencing and machine learning technology with breakthrough research on the forest microbiome to put the right native, biodiverse communities of mycorrhizal fungi in the right place. This leads to more quality wood created more quickly, more carbon sequestered, and more resilient forests. We serve land partners seeking to implement regenerative forestry practices that improve productivity and sustainability, and corporate carbon buyers building high-quality carbon removal portfolios with strong social and biodiversity outcomes.

Our analytics platform identifies forests that harbor wild, biodiverse and growth promoting communities of fungi to develop naturally sourced inoculants. We are building fundamentally new ways to restore forest fungal biodiversity to create climate impact.

The Challenges Funga Faced

As Funga scaled up its efforts, a number of operational challenges began to emerge which not only consumed their valuable time but also impacted the accuracy and traceability—two critical elements for a science-based organization.

• Slow Workflow Execution: Their R-based bioinformatics setup required approximately 4 hours per run, hindering rapid iteration.
• No Schema Standardization: Due to inconsistent table structures, covariates had to be recalculated for each data source.
• Redundant Reprocessing: Even minor changes required full workflow re-execution because selective task execution wasn’t supported.
• Manual Data Merging: Biodiversity, soil, and sequence data were manually combined, increasing the risk of human error.
• Missing QA Framework: There were no dynamic data validations or automated flags to identify problematic records in delta tables.

How Koantek Responded

To address these limitations, Funga partnered with Koantek, a leading provider of modern data and AI solutions. The joint effort focused on re-architecting Funga’s workflows to enhance scalability and automation as well as implementing a robust QA/QC module process to enhance data quality, operational efficiency and governance. 

Here’s how Koantek transformed Funga’s bioinformatics infrastructure:

• Automated Workflows: Replaced R-based processes with Python pipelines, reducing workflow runtime from 4 hours to 15 minutes and pipeline initialization from 20 mins to 20 seconds,  boosting scalability.
• Built-In QA/QC: Implemented YAML-driven validations to flag data issues early and ensure traceable, high-quality outputs.
• Scalable Architecture: Enabled schema evolution and source-specific configs, eliminating redundant processing and manual recalculations.
• Governance and Insights: Integrated Unity Catalog for data lineage, access control, and real-time dashboards to track KPIs and data health.

The Outcome

The outcome of Koantek’s engagement with Funga was transformational—not only in terms of technical performance but also operational agility and contributing to the advancement of scientific biodiverse output.

• 90% Faster Workflow Processing: Setup and execution time were slashed from nearly 4 hours to just 15 minutes.
• 60x Faster Setup: Pipeline initialization now takes only 20 seconds—down from 20 minutes.
• 30k+ Data Issues Flagged: Automated QA/QC surfaced over 30K potential issues, enabling faster resolution by field and data teams.
• Improved Data Integrity: Validations built into the pipeline now ensure cleaner, more trustworthy datasets.
• Reduced Manual Intervention: Automated workflows eliminated the need for repeated reprocessing and manual data merges.
• Real-Time Monitoring: Dashboards provide continuous visibility into critical KPIs and operational metrics.
• Efficient Covariance Handling: Modular architecture removed the need to recreate output tables, saving time and reducing compute overhead.
• Cost Optimization: Optimized resource usage and parallelized execution led to significant infrastructure savings.

‍

“Our experience with Koantek has been excellent. Their team demonstrated a deep understanding of our requirements and executed the project with efficiency and professionalism. They provided invaluable support in enhancing our Databricks pipelines, integrating robust QA/QC mechanisms, automating environmental data integration, improving data traceability, optimizing workflow by 90%, cost optimization and operation efficiency. Communication was clear throughout the process, and they were proactive in addressing challenges. Their expertise and collaborative approach helped streamline our workflows, ensuring scalability and seamless integration with Unity Catalog. Strong collaboration and expertise from the team. We appreciate the patience, kindness, and professionalism demonstrated throughout this work and are very pleased with the outcomes.  We would highly recommend Koantek for future collaboration as we continue to scale our data infrastructure.”