Praful Saklani

Co-Founder and CEO Pramata

Last Reviewed: September 19, 2026

Executive Summary

Praful Saklani co-founded Pramata and runs it as chief executive. The company works on a problem large enterprises consistently underestimate: they sign thousands of contracts, then lose track of what those contracts actually oblige and entitle them to. Price escalators go unenforced, renewal windows pass, vendor terms drift out of line with what was negotiated, and compliance obligations sit unread in documents nobody has opened since signature. Pramata extracts and structures those commercial terms so they can be acted on, selling to enterprises including McKesson, Intercontinental Exchange, AbbVie and Callaway Golf. Saklani came to it after an unusually varied run — an AI and internet company in corporate travel, a consultancy he co-founded and ran, and an executive role at a social enterprise supplying affordable water purification in developing countries. He is a consistent public sceptic of enterprise AI hype, arguing in 2026 that AI does not behave like conventional software and should not be procured or deployed as if it does.

Career Highlights

• Co-founded Pramata and has led it as chief executive throughout, building an enterprise contract intelligence platform.
• Won enterprise customers including McKesson, Intercontinental Exchange, AbbVie and Callaway Golf, addressing revenue protection, vendor cost reduction, contract lifecycle acceleration and compliance.
• Founded and led Yatra Corporation, which applied artificial intelligence and internet technology to corporate travel management optimisation.
• Co-founded the consulting firm Invotech Systems and served as its managing partner.
• Held an executive role at WaterHealth International, a social enterprise delivering affordable water purification to developing countries.
• Publishes regularly on the practical limits of enterprise AI, including 2026 pieces on why “good enough” models often beat the most powerful ones and what it takes for an AI agent to be production-ready.
• Graduated from Swarthmore College with High Honors in economics and political science.

Professional Journey

Saklani read economics and political science at Swarthmore College, graduating with High Honors — a social science rather than engineering background, which is visible in how he frames Pramata’s product as a commercial and legal problem rather than a document-processing one.

His first company was Yatra Corporation, which he founded and led, applying artificial intelligence and internet technologies to corporate travel management. Attempting machine learning against a messy, rules-heavy commercial domain is essentially the same technical problem he would return to later with contracts: the data exists, it is unstructured, and the value lies in extracting decisions from it rather than in the model itself. He also co-founded Invotech Systems, a consulting firm, where he was managing partner.

Between those ventures he held an executive position at WaterHealth International, a social enterprise that builds and operates affordable water purification systems for communities in developing countries — operationally demanding work in a sector with no software margins, and an unusual entry in a software founder’s history.

Pramata followed. The premise is that contract lifecycle management tools focus on getting agreements signed, while the commercial value sits on the other side of signature, in the accumulated estate of contracts already in force. Pramata positions itself as contract intelligence rather than contract management: extracting the negotiated terms from existing agreements at scale and making them queryable, so that finance can find unbilled entitlements, procurement can see where vendor terms have drifted, and legal can answer compliance questions without re-reading the paper.

That focus has aged well against the current wave of generative AI, and Saklani’s public writing in 2026 reflects a practitioner’s caution about it. His argument is that AI does not work like software and should not be treated like it — that probabilistic systems need different procurement, evaluation and production standards — and that for many enterprise use cases a smaller, cheaper, adequately accurate model outperforms the most capable one available, because the binding constraints are cost, latency and predictability rather than raw capability.

Education

Swarthmore College

Bachelor’s degree with High Honors, Economics and Political Science