How Birlasoft turned contract obligations into a delivery advantage with AI agents and Microsoft Azure

|
3 minutes
Graphic representation

How do you keep track of hundreds of customer commitments buried across multiple contracts? As contracts become increasingly complex, obligations could be scattered across master service agreements (MSAs), statements of work (SOWs), milestone schedules, service-level commitments and reporting requirements. Missing even one could affect delivery performance, customer trust and financial outcomes.

When global technology firm Birlasoft decided to build a solution with AI agents, Microsoft Azure emerged as the natural choice.

“We wanted a solution that could scale, bring us the security aspects required and provide predictability in the overall execution,” says Ramasubramanian S, Senior Vice President, Delivery Head-BFSI, Birlasoft. “When we assessed the options, we felt Azure could bring us everything we were looking for from a solution standpoint,” he adds.

The result is a multi-agent Contract.ai solution that helps delivery teams start projects with greater clarity, accountability and confidence.

Promise: Make every commitment visible 

Before Contract.ai, the handover from contract signature to project execution depended on an intensive manual process. MSAs, SOWs, task orders, and other contractual documents were shaped by legal, finance, taxation, sales and delivery teams. Once signed, they passed to delivery managers, who had to interpret legal language, operational commitments and validate it with multiple stakeholders.

“We realized the manual process, even if I brought in a maker-checker (two-person approval) or a couple of more reviewers, was not going to be foolproof,” Ramasubramanian says.

Ramasubramanian S, Senior Vice President, Delivery Head-BFSI, Birlasoft. Photo by Ramasubramanian S.

Birlasoft needed a way to create a reliable bridge between the agreements it signed and the actions delivery teams were expected to execute. But solving the problem required more than automation. The objective was to make every commitment visible, connect it to the right stakeholders and create a mechanism for continuously tracking whether it had been fulfilled.

Proof: A coordinated system of humans and agents

“When we looked at the contracts, we realized one agent could not address every nuance,” Ramasubramanian says. “We wanted to build agents that could cater to the specific requirements of different contract types.”

The system leverages Azure Data Factory for ingestion, Azure Data Lake for data storage, Azure Databricks for analytics and reporting, Azure Key Vault for access management and Azure AI Content Safety to support responsible AI practices. Using 335 metadata elements, the solution interprets contractual context, extracts obligations and links every requirement back to its original source.

Today, six specialist agents, orchestrated by a planning agent, work together to extract obligations, validate outputs and flag inconsistencies. However, human oversight remains central to the process. “These are obligations that have financial and legal implications, and we cannot let the agents run rogue on this,” Ramasubramanian says.

The proof is also in what the system can connect. Ramasubramanian recalls how Contract.ai linked language on page two of a document to an obligation referenced on page 25. “It’s something very difficult for a person reading sequentially to spot it so quickly,” he says. 

Productivity: Time, risk and accountability

Today, Contract.ai is delivering a realized annual benefit of approximately 4,000 person-hours saved, processes more than 1,000 SOWs a year and identifies an average of more than 10 obligations per project.  As the company grows, the number of SOWs, delivery managers and hours saved is increasing. But Ramasubramanian sees productivity as only one dimension of the result. 

“The biggest takeaway is that I can be comfortable that I am not missing an obligation,” he says. “That has far-reaching consequences in terms of the financials and my reputation as a strong delivery unit.”  The deeper benefit is confidence. Commitments are captured, assigned, tracked and reviewed before they become risks.

A man wearing formal clothes and spectacles

Frontier Employee

“AI-enabled Obligation Tracking System with Contract.ai has enabled us to focus more on delivery execution rather than contract interpretation. The real-time visibility into due, overdue, and completed obligations has strengthened governance, improved risk management and enhanced our ability to proactively meet customer commitments.”
Vimal Kant, Programme Manager, Birlasoft

English (United States)
Your Privacy Choices Opt-Out Icon Your Privacy Choices
Consumer Health Privacy Sitemap Contact Microsoft Privacy Manage cookies Terms of use Trademarks Safety & eco Recycling About our ads