AI-Augmented SDLC with Jason Adams of State Street Alpha & Charles River Development; Accelerating Software Development at Enterprise Scale
CIO Classified
Limits of AI for full autonomous recommendations
Jason cautions LLMs struggle to satisfy deep non-functional constraints for production-grade deployments and operations.
When you’re managing $60 trillion in assets across dozens of products and 30 global jurisdictions, technical debt isn’t just an inconvenience—it’s an existential risk.
Jason Adams, Interim CTO of Charles River, a State Street Company, leads 800 engineers building mission-critical trading platforms for the world’s largest asset managers. Joined by Sid Pardeshi, Co-Founder and CTO of Blitzy, he explains how State Street is using an AI-augmented SDLC to modernize decades-old systems, refactor legacy code, and dramatically increase developer productivity—without compromising the rigor required in financial services.
Jason frames the strategy around three pillars:
- AI for engineering (copilots and polyglot support),
- AI for operations (APM, observability, and proactive monitoring), and
- AI embedded in products (LLM-powered explainers).
Using Blitzy’s agentic approach—iterative context building, dependency mapping, and targeted code generation—State Street compressed months of work into weeks while maintaining strict quality gates.
About the Guests:
Jason Adams
Jason Adams is the Interim CTO of Charles River, a State Street Company. He brings deep expertise in modernizing legacy fintech infrastructure into scalable, cloud-native systems that support mission-critical financial services at global scale.
Previously, Jason was Head of Platform Product and Strategy at Charles River Development and CTO of Mercatus (acquired by State Street and now part of Charles River for Private Markets). He has led high-impact initiatives across engineering, product, and cloud infrastructure, with extensive experience guiding end-to-end delivery teams.
Today, Jason is driving a comprehensive SaaS transformation at CRD, focused on building resilient, future-ready architectures. From scaling global engineering organizations to delivering secure, high-performance platforms, he is committed to advancing innovation, agility, and long-term growth across Charles River, State Street Alpha, and State Street.
Sid Pardeshi
Sid Pardeshi is a technology leader and entrepreneur, currently Co-Founder and CTO of Blitzy. He holds a Harvard MS/MBA and previously served as a Software Architect at NVIDIA, where he built deep expertise at the intersection of AI, large-scale software systems, and product innovation.
At NVIDIA, Sid was recognized as a Master Inventor, earning the Inventor’s Jacket for driving AI-powered product innovation, with more than 25 U.S. patents filed across gaming, augmented reality, and virtual reality. He is also a seasoned software engineer with a strong track record in application performance optimization, delivering native client load-time improvements of up to 90%.
Beyond hands-on engineering, Sid has led and coordinated software design, framework requirements, and application architecture across global teams of 500+ engineers. Today, he applies this blend of innovation, technical depth, and organizational leadership to building autonomous software development platforms that help enterprises modernize at scale.
Timestamps:
00:30 – Jason on Managing $60 Trillion in Assets
01:55 – Challenges and Strategies in Financial Services
07:00 – Embracing AI for Modernization
09:10 – AI in Software Development Lifecycle
15:55 – Ensuring Quality and Compliance with AI
23:55 – AI in Operations and Incident Response
26:00 – Proactive Workflow Monitoring
26:20 – AI in SDLC: Creation to Operations
30:00 – Challenges in AI Recommendations
33:20 – Iterative Context Building with AI
36:00 – Human Side of AI Transformation
42:30 – Adopting AI Tools in Financial Services
Guest Highlights:
"One of the things that excites me the most right now is the ability to use an AI-augmented SDLC to drive modernization. Otherwise, with this many systems, it's too hard." — Jason
"You have to invest in the non-attractive parts first. You have to build a foundation that's gonna support being able to bring on solutions and tools that could change your overall enterprise SDLC. That's a lot of work and that's a major investment." — Jason
"We are unlocking by adding these additional capabilities and additional assurance that improves quality exponentially more than we could have in the past. Now I can have an agent swarm check itself—multiple agents doing code review at a level of depth we just don't have time to get to." — Jason
Get Connected:
Hungry for more tech talk? Check out latest episodes at ciopod.com:
- Ep 63 - How Autonomous AI is Solving the Enterprise Modernization Challenge
- Ep 62 - Running IT Like a Growth Engine
- Ep 61 - What Manufacturing Can Teach You About Scaling Enterprise AI
Learn more about Caspian Studios: caspianstudios.com
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
The AI-powered Podcast Player
