Entrepreneur / CEO / COO / General Manager building great technology product and services businesses in cloud, enterprise software, SaaS and related products/services. Passionate early-adopter and evangelist for cloud computing - and all things cloud-native -  and its ability to transform IT. Strong across multiple competencies - sales, marketing, technology, finance & operations.

John. Treadway

I believe there’s a parallel between the autonomous driving market and internal AI strategy. Let me know what you think.

Elon Musk has said he doesn’t understand why other automakers won’t license Full Self-Driving (FSD) from Tesla. Some automakers, notably Rivian, Ford, and Mercedes-Benz, have determined that this technology is so central to their businesses that they cannot just abdicate it and let another company control their destiny. While this makes for more work and higher upfront risk, they are not wrong to want to "own" something so fundamental.

In 2026, we are seeing a similar shift in the enterprise. The bigger companies are often avoiding "renting" external AI stacks tied to a single vendor (OpenAI, Salesforce, etc.) in favor of building a sovereign internal AI stack. Here is why the "Own vs. Rent" debate is shifting:

- Data Sovereignty & Security: You keep your IP and customer data behind your own firewall, not a third-party black box.

- Operational Control: You aren't at the mercy of a vendor's pricing hikes, model deprecations, or roadmap pivots. Control your roadmap, control your destiny.

- The "Unified Brain": An internal stack connects all your siloed data (CRM, ERP, Slack) into one cohesive intelligence layer.

- Unit Economics: While "renting" is cheap at first, building your own infrastructure avoids the compounding "SaaS tax" at scale. If you have 2,000 employees, even part of the stack will be $720k per year (at $30). Add in the siloed AI add-ons from Salesforce, ServiceNow and others and that number can quickly balloon! You can "own" your own stack for a lot less - and use the savings to fund enablement, process change and other super-important efforts. 

- Custom Moats: Generalist vendor AIs are a commodity; an AI stack fine-tuned on your proprietary processes is a competitive advantage.

The takeaway: If AI is going to be the "engine" of your company, do you want to own the engine, or just lease the same system your competitor is using?

#AIStrategy #EnterpriseAI #SovereignAI #Innovation #DigitalTransformation

John Treadway

CEO of AI Technology Partners, provider of governed, adopted, enterprise AI—deployed in our client’s environment and under their control

I am very excited to welcome Dave MacKay as SVP, Client Strategy & Partnerships at AI Technology Partners!

Dave is a senior go-to-market and strategy leader with more than 25 years of experience helping technology companies translate AI, advanced analytics, and cloud capabilities into clear customer demand and measurable business outcomes. At AI Technology Partners (AITP), he is focused on accelerating early customer traction and shaping our go-to-market strategy in the generative and agentic AI landscape.

His background spans enterprise analytics, cloud platforms, and professional services, with leadership roles at SAS, Cognos, Onica, and Slalom, as well as early-stage growth firms. Across these environments, Dave has consistently helped organizations open new markets, validate repeatable sales motions, and extend platforms into new use cases and verticals. He is particularly known for building and leading high-performing, cross-functional teams that align strategy, sales, partners, and delivery around concrete outcomes.

Dave’s strength lies in bridging strategy and execution. He brings a pragmatic, market-driven perspective that helps teams focus on what actually drives adoption, revenue, and scale.

Dave is working closely with me and the rest of the leadership team to inject real-world market signal into how we position, sell, and partner as the company grows. His mandate is to drive early market traction through hands-on collaboration with clients and strategic partners, helping refine and scale AITP’s go-to-market approach.

Outside of work, Dave enjoys staying active and spending time outdoors, including fishing and tennis.

Dave is widely respected for his strategic judgment, collaborative leadership style, and ability to create clarity through action. We are excited to have him join AITP as we continue to build practical AI solutions that deliver real business outcomes.

DocRouter

Spearheading DocRouter.AI, a cutting-edge AI Backbone for Document Workflows, enabling companies to implement AI workflows with human-in-the-loop for precise processing of unstructured information.

DocRouter has an open source core, and an Enterprise, on-VPC version - accelerating AI adoption for various industry verticals and reducing processing time by 90% while eliminating manual data entry errors.

Founder

Analytiq Hub

+ Dynamic and results-driven consultant with 20+ years of engineering experience
+ Proven track record in leading teams, managing projects from start to finish, and delivering innovative solutions
+ Proficient in client relationship management, strategic planning, and subcontractor procurement
+ Excels in fostering collaborative  environments to maximize team performance and exceed client expectations

I regularly give talks about robotics, language models and other tech: https://analytiqhub.com/talks/

Projects completed and ongoing:
+ AI for Digital Health, FinTech
+ AI workflows for administrative and revenue cycle tasks
+ Chat, summarization, retrieval-augmented generation (RAG)
+ RPA and IPA automation
+ Business rules engines
+ DevOps, MLOps
+ Business Intelligence
+ Software tools include Databricks, Snowflake, AWS, Azure, GCP, OpenAI, LangChain,  John Snow Labs, HuggingFace, PyTorch, Python React, Tailwind, NextJS, ROS, C/C++, Linux kernel & drivers, Postgres, MySql, MongoDB

isee ai

+ Architect ML+Data infrastructure
+ Design MLOps and Data Lakehouse using Databricks and AWS
+ Partnerships with cloud vendors, work with contractors
+ Integration with planner, simulation, controls, using C++, Python, ROS, PyTorch, online and offline, lidars, GPS and other sensors, in a high-performance and real-time environment.

Staff Software Engineer

HPR (Hyannis Port Research) 

+ Translate broker-dealer market trading requirements into product design, implementation, and release delivery
+ Develop and integrate components of Softbot (Market Access Gateway) with broker-dealer order management system for dark pools and lit venues
+ Design and implement AMPS pub/sub protocol interface for high performance, low latency market access
+ Python based Order Management System for integration testing
+ FPGA based Linux network driver for Riskbot and Omnibot (ultra low latency pre-trade Risk and Market Access Gateways)
+ cron application for automating component bringup/shutdown
+ rpm installer for HPR software suite, automated deployment through yum repositories
+ setup of gitlab pipeline
+ make, cmake development

Senior Software Engineer

Bitdribble

EXFO

+ Lead designer and implementer of the EVA, Virtual Verifier, BV3000, BV3100, BV1100, BV1500, BV4200, BV4220, BV4230 Verifier platforms
+ Platform libraries, CLI, upgrade mechanism, microsecond accurate timestamp support, NTP, GPS, CDMA, 1588 for time synchronization, debug tools, diagnostics
+ Test suite development for Linux verifier platforms
+ New communication protocol for verifier-to-server communication (comm layer)
+ Functional blocks: plugin-testapp, registry-client used by Web Verifier and EXFO Mobile Agent (EMA)
+ CDMA wireless dongle support for Linux verifiers
+ Openssh, openssl code integration; ssl related CLI commands
+ Postgres, FreeRadius, xerces, expat open source code integration
+ net-snmp open source code integration; snmp-agent code block; SNMP related CLI commands
+ IPv6 support on all Linux verifier platforms
+ Stack trace and core file reporting to allow debugging in the field
+ RAID support 
+ Local test results storage on postgres database support; db-agent implementation
+ Cross compilation for Android, Raspberry Pi, OpenWRT
+ Cloudinit support
+ OpenStack, VmWare, Xen, Microsoft Hyper-V hypervisor support
+ RPM, Debian, Docker installers
+ Parallel make support
+ Conversion from CVS to Git

Linux kernel 2.6.16, 2.6.35.7, 3.18.12:
+ IP stack amended to pass up hardware- and driver-level timestamps. 
+ Socket and interface level Vlan support
+ Multiple new drivers for Ethernet, Serial, I2C 

Test development:
+ VAST test allowing scripting on tests on verifier probes (patent pending)
+ Ping, Traceroute
+ Peer to peer library implementation
+ Build tools, unit testers

Software Architect

Brix Networks

+ Lead designer and implementer of the Brix 3500, 4000, 4100 and 4104 Verifier platforms.
+ Enhancements and refactoring of code for OSE-based Brix100, Brix1000, Brix2500 platforms
+ Test suite development for the Linux and the OSE-based verifier platforms. 

OSE application code
+ Ping, Traceroute, H.323 test suite implementation
+ Libraries for ntp, IP stack config, logger, sem-server 
+ Open source ports to OSE: ntp, openssh, openssl
+ Vendor code ports to OSE: Radvision stack
+ Leak-tracking functionality added to OSE heap
+ OSE verifier CLI commands for interface configuration, logging

OSE IP stack work:
+ Support for alias interfaces, vlan, timestamping, routing support for interface groups 

OSE driver work:
+ PHY-level timestamping mechanism in Ethernet driver on Brix100 platform; refactoring and optimizing Brix1000, Brix2500 Ethernet drivers.

Portability layer for OSE, Linux, Windows code:
+ Symbol lookup mechanism for OSE plugins in main application
+ Implementation and refactoring of mutex, event, thread abstractions for OSE, Linux, Windows
+ Portable arithmetic routines
+ DNS resolver, IP address format conversion, portable wallclock, time format conversion

Portable test infrastructure (Linux, Windows, OSE):
+ Portable test manager, plugin manager, action manager
+ Queuing of test results to ensure results are not dropped if network temporarily goes down

Linux verifier infrastructure code:
+ Configuration manager functional block
+ CLI
+ Integration of libc libraries and other supported libraries
+ Integration of busybox as binutils
+ Communication to server
+ Image upgrades

Makefile work:
+ Implementation and refactoring of central makefile infrastructure for verifier platforms and for the test suite

Principal Software Engineer

PictureTel

Member of PictureTel’s H.323 group, which developed the H.323 protocol stack used by the PictureTel video conference products. 
+ Implemented Fast Start and H.245 Tunneling support.
+ Implemented Gatekeeper Discovery Support.
+ Bug fixes and enhancements in all layers of PictureTel’s H.323 stack.

Software Engineer

Massachusetts Institute of Technology

- PhD, Math (Homotopy Theory)
- Teaching assistant, Calculus 18.01-18.03, Linear Algebra 18.700, Intro to Topology 18.901

Teaching Assistant

━━
🚀 DocRouter.AI
━━

I am the founder of DocRouter.AI, an open-source platform revolutionizing document processing through AI-driven automation. 

DocRouter.AI automates data extraction while incorporating a human-in-the-loop design to ensure financial accuracy. This approach cuts processing time by up to 90% and eliminates manual data entry errors. ​Key features:
→ Autonomous processing with AI/LLMs
→ Human-in-the-loop design for financial accuracy
→ Direct ERP integration capabilities
→ REST APIs for all functions
→ Open-source​

Use cases span across industries, including clinical trial invoice processing, shipping container manifest processing, and financial report analysis. ​

━━
🧠 Additional Ventures and Expertise
━━

In 2023, I founded AnalytiqHub to simplify the adoption of cutting-edge data and AI techniques for digital health and fintech enterprises. Our projects have encompassed:
→ Clinical workflows enhanced by AI
→ Revenue Cycle Management utilizing Large Language Models
→ Robotics and Computer Vision applications​

With over 20 years of software engineering experience, I have:
→ Implemented full-stack AI solutions in fintech and health tech
→ Architected robotics and AI infrastructure for self-driving cars
→ Built high-performance stock market trading infrastructure for Wall Street
→ Worked extensively with embedded software, hardware, FPGAs, Linux kernel, and computer networks​

━━
🎓 Community Engagement and Mentorship
━━

I am actively involved in the Boston tech community, serving as a speaker at industry events and expert panels on topics such as LLMs, self-driving cars, and entrepreneurship. My commitment to mentoring includes working with early-stage startups and college students on innovative projects. Notable engagements include:
→ 2024 Hack MIT panel judge
→ AI Furnace invited panelist
→ AI Camp guest speaker and organizer​

━━
🏫 Civic Involvement
━━

Residing in Lexington, MA, I have served as a Town Meeting member and previously on the Appropriation Committee, gaining deep familiarity with municipal and budget issues. I created lex-wiki.org, a local wiki for Lexington, and organized the Candidate Questionnaire for Lexington elections between 2014-2018. I mentored Lexington High School students exploring Generative AI applications for interacting with public records and town budget data.​

I hold a PhD in Mathematics from MIT and am a founding member of smaht.ai.​

Feel free to connect with me to discuss AI-driven document processing, innovative applications of LLMs, or community tech initiatives.

taplio

The Battle

John. Treadway

Andrei Radulescu-Banu

Followers

Posts

Engagements

Likes

Performance ScoreJohn. Treadway's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

John. Treadway

Andrei Radulescu-Banu

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}John. Treadway's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreJohn. Treadway's LinkedIn Performance