Taller Top 10

Ten of the strongest projects Taller has delivered, with their published results.

1

adidas

Retail, Commerce & Consumer Brands

RPA UiPath: Back-Office Efficiency and Data Integrity at Scale

Problem
Adidas needed to improve back-office efficiency and data quality at a scale where manual entry both consumed expensive headcount and introduced reconciliation defects. RPA alone does not solve the problem; instead, the underlying data substrate has to be hardened simultaneously, or the automation amplifies existing defects rather than removing them.
What Taller did
Taller's RPA team of developers and data engineers implemented UiPath automations across finance and operations workflows while parallel work hardened the underlying SQL Server and ETL processes for performance and security. UiPath Studio and Robot deliver the automation surface; SQL Server tuning, ETL refactoring, and integration security harden the substrate the bots run against.
Results
The RPA program produced more than $2M in annual cost savings.
Read the case study on taller.ai
2

Envision Healthcare

Healthcare & Life Sciences

RPA Accelerator and Nearshore Replacement of Onshore Talent

Problem
The strategic move behind the broader Envision engagement was a structural cost reduction: replacing onshore IT with senior nearshore talent across RPA, data engineering, and the supporting roles. Nearshore replacement of onshore IT only works when the talent has the seniority and the domain familiarity to operate without the institutional knowledge the offboarded onshore team is taking with them.
What Taller did
Taller rapidly deployed six senior consultants in the first wave, Lead, Senior, and Mid RPA developers; a senior data engineer; a Snowflake data engineer; a Progress OpenEdge developer, and expanded into additional roles as the accelerator program absorbed more processes. The Lead and Senior RPA Developer roles are the operational insurance policy: they hold the methodology while the rest of the team ramps, and they are the ones who interface with Envision's business stakeholders during the transition.
Results
The accelerator program delivered a thirty-percent cost reduction relative to the onshore baseline and a nine-percent reduction in bot-related incidents, earned through the engineering-best-practices discipline Taller's senior team brought to the existing bot estate.
Read the case study on taller.ai
3

Greenlight

Financial Services & Fintech

Cost Optimization: Optimizely Rolling Fingerprint Refactor

Problem
A core function in the Optimizely service maintains a "rolling fingerprint" at the user-by-feature level, writing a record to a DynamoDB table for every user and feature key combination. At peak, this generated up to 1,500 writes per second, driving significant and largely unnoticed DynamoDB costs amounting to over $200K per year.
What Taller did
The engagement began with a deep investigation into the original implementation, tracing how the rolling fingerprint logic worked, instrumenting and tracking usage metrics to quantify actual write volume, and identifying where writes could be reduced without affecting functionality. The team then executed a refactor to lower the number of DynamoDB writes required to maintain the fingerprint, followed by thorough QA to validate that behavior remained correct.
Results
Direct cost reduction. The refactor cut DynamoDB write volume against the fingerprinting table, delivering savings of over $200K per year on a previously unmonitored expense. Cost visibility.
Read the case study on taller.ai
4

Zscaler

Cybersecurity

ZIA Page Modernization with Chiron

Problem
Zscaler needed to modernize legacy ZIA user-interface pages into its newer frontend architecture. The work required more than a visual rewrite.
What Taller did
Taller used Chiron to orchestrate the modernization project as a hybrid human-agent workflow. The Knowledge Database gave the Chiron agents access to the relevant technical context: page scope, frontend patterns, reference implementations, Nimbus component usage, coding guidelines, dependency expectations, and modernization requirements.
Results
Using Chiron, Taller migrated three ZIA pages in two days. By comparison, Zscaler had previously spent approximately 14 days to migrate one page without AI, and approximately five days to migrate one page using AI platforms such as Devin. The Chiron-based approach demonstrated a step-change in modernization throughput: three pages in two days, with the work structured around reusable components, shared patterns, quality artifacts, and readiness for integration into the modernized ZIA frontend architecture.
Read the case study on taller.ai
5

United Airlines

Aviation & Logistics

RAG for Aircraft Manual Comparison: Compliance and Version Control

Problem
Aircraft maintenance manuals at a major airline run into hundreds of PDF and XML documents, each updated by manufacturers on irregular cadences. Identifying critical changes between versions had been a manual page-by-page review process, taking weeks of skilled time and a non-trivial source of regulatory exposure when changes were missed.
What Taller did
Taller implemented the retrieval-augmented system that compresses version comparison into minutes of automated analysis with full traceability of every detected change. The architecture decomposes each manual into the semantic chunks compliance reviewers actually care about, procedures, parts numbers, torque specifications, safety callouts, and compares versions at the chunk level.
Results
The deployment reduced manual document processing time from weeks to minutes, improved accuracy in identifying critical version changes, and delivered full traceability of changes for regulatory compliance.
Read the case study on taller.ai
6

United Airlines

Aviation & Logistics

Customer Care Analytics: Centralized Self-Service Insight for MileagePlus

Problem
The customer-care function at United faced the familiar operational economics problem: a high volume of repetitive member inquiries that did not require an agent's judgment but consumed agent capacity. Without a self-service substrate, every additional MileagePlus member converted directly into incremental contact-center cost.
What Taller did
A four-person Taller team stood up a Tibco Spotfire-based analytics workflow in under three weeks, centralizing customer-care data that had previously been distributed across operational systems. The team built Python and SQL pipelines against Oracle and SQL Server sources to push normalized event data into Spotfire's analytical layer.
Results
The analytics workflow delivered a thirty percent reduction in agent dependency through self-service across the affected categories, with a significant decrease in call volume via chat-based case creation.
Read the case study on taller.ai
7

Kforce

Professional Services & Staffing

AI for Job-Description Classification: Scraping, Modeling, Extracting at Scale

Problem
Kforce's competitive intelligence function wanted to know which companies were posting job descriptions on staffing-agency sites under generic descriptors: an intelligence-gathering exercise that would let the firm sharpen its competitive positioning across technology and financial-services placements. The challenge was building the data pipeline and the classification models at production scale rather than as a research notebook.
What Taller did
Taller built and trained the AI models against scraped job-description content, categorizing roles by sector and extracting the skills each description named. Scrapy and Selenium handle the harvesting layer; Pandas and Jupyter Notebook hold the data-science iteration cycle; TensorFlow supports the model-training workload.
Results
Taller's models delivered a sixty-percent improvement in data-extraction efficiency, a thirty-five-percent boost in talent-conversion rate, and a seventy-percent increase in market-analysis speed.
Read the case study on taller.ai
8

Kforce

Professional Services & Staffing

Taxonomy Unification AI: ML, Deep Learning, and NLP Across Recruiting Data

Problem
Inconsistent data taxonomies across the recruiting platforms the firm operates were running into Kforce's AI ambitions repeatedly. Even the best matching algorithm produces low-confidence outputs when the source data is wrong in the slow ways that compound across systems; same role title meaning different things across employer contexts, same skill described with different vocabulary at different career levels.
What Taller did
Taller deployed the integrated ML, deep-learning, and NLP solution that unifies and optimizes the taxonomies, centralizes the data, and enables accurate job-order matching and candidate probability scoring at production scale. The Python and SQL pipeline handles the data extraction and normalization rules that catch the easy variants; the deep-learning model handles the harder cases.
Results
Taller's solution produced a fifty-percent reduction in data discrepancies, a thirty-percent improvement in candidate-matching accuracy, and a forty-percent increase in recruiting efficiency.
Read the case study on taller.ai
9

Boost Innovations

Healthcare & Life Sciences

Agentic Healthcare Scheduling and Provider-Intelligence Automation

Problem
Healthcare operations involve constant scheduling complexity. Appointments get canceled, patients need to be rescheduled, providers have different specialties and availability, doctors work across different schedules, and administrative teams need to match the right patient with the right provider at the right time.
What Taller did
Taller used Chiron to build an AI agentic automation integrated with the Boost Results Platform. The agent can retrieve and reason over platform information related to patients, doctors, providers, appointments, schedules, specialties, and availability.
Results
Using Chiron, Taller compressed a project that would typically take six months into one month. Boost now has an agentic automation capability that can be used in client demos to show how AI can retrieve healthcare platform data, answer operational questions, analyze providers and schedules, and support appointment cancellation and rescheduling workflows. The agent can reason across patients, doctors, providers, specialties, schedules, and availability, turning static platform data into an interactive operational assistant.
Read the case study on taller.ai
10

Milwaukee Tool

Industrial, Manufacturing & Supply Chain

Global Supply Chain Modernization: Oracle Integrations

Problem
As Milwaukee Electric Tool expanded exponentially, scaling global sales from $500M toward a $10B trajectory, the enterprise rapidly outgrew its fragmented, multi-system legacy infrastructure. To streamline operations across more than twenty global manufacturing and distribution facilities, the company initiated an enterprise-wide go-live migration to Oracle Fusion Cloud ERP and SCM.
What Taller did
To bridge this transition, a multi-vendor delivery pod composed of full-time employees (FTEs) and global onshore/offshore contractors was stood up to replace the outgoing implementation teams and absorb the footprint left behind by Accenture. Operating under the direction of Product Owner Claire, engineering resources immediately prioritized reverse-engineering the undocumented environment, progressively gaining full system authority.
Results
Uninterrupted Supply Chain Continuity: Successfully assumed one-hundred-percent operational ownership of Oracle's footprint during a high-risk vendor transition period (Accenture hand-off) without experiencing facility or manufacturing downtime. Production API Delivery: Engineered and launched a real-time Oracle inventory-query API, significantly accelerating supply chain data visibility. Drastic Agile Cycle Optimization: Successfully rebalanced team meeting metrics, enabling leaders to systematically strip out bloated calendar invites and reclaim engineering time.
Read the case study on taller.ai