Solution Architect (Senior AI / Technical Lead)

Bits In Glass (BIG) is a high-growth AI and automation consulting firm with offices across Canada, the United States, India, and the United Kingdom. Our portfolio spans more than 10 leading technologies across AI, cloud, data, automation, and business applications.
Recognized globally as a Great Place to Work and the recipient of multiple partner excellence awards, BIG is built on collaboration, innovation, and delivering measurable business outcomes for enterprise clients. We are a team of experienced technology professionals who enjoy solving complex challenges, celebrating wins together, and helping customers modernize with confidence.
AN EXCITING OPPORTUNITY WITH BIG AND CONVENE AI
BIG is excited to be partnering with Convene AI, an AI-focused technology company developing innovative solutions that leverage artificial intelligence to solve real-world business challenges.
As part of this partnership, we are looking for talented professionals to join BIG and work directly on Convene AI projects and initiatives. You will initially be hired as an employee of Bits In Glass, while working closely with the Convene AI team and contributing to the development and growth of their AI solutions.
The role is expected to remain on BIG’s payroll for approximately 12–18 months, with the intention of transitioning to a permanent position directly with Convene AI following this period, subject to business requirements and transition arrangements.This is an exciting opportunity to contribute to an emerging AI organization at an important stage of its growth, while benefiting from the support, culture, and established consulting expertise of Bits In Glass.
ABOUT THE ROLE
This is the highest-leverage technical role at ConveneAI. Every architectural decision you make is inherited by multiple client engagements, and the accelerators and standards you build determine how fast the next pod moves.
The mandate has three parts. You shape engagements before they are sold, by assessing what is genuinely buildable in 8 to 12 weeks and saying so honestly.
You govern architecture across live pods, making the decisions a single embedded engineer should not make alone: model hosting, data residency, security posture, integration patterns. And you lead the technical craft, through design review, mentorship, and the reference architecture that every pod starts from.
You will remain hands-on. You will read code, prototype when it is faster than arguing, and stand in front of a bank's or an insurer's architecture review board and defend a design under real scrutiny. If you have drifted into slideware architecture, this will not suit you.
WHAT YOU'LL DO
Shaping and Pre-Sales

• Join client conversations during the sales cycle to assess technical feasibility, and scope what can realistically reach production in 8 to 12 weeks.

• Push back on requirements that cannot be delivered, before they become contractual commitments on a fixed-fee or outcome-based engagement.

• Own the architecture decisions that determine engagement profitability: model selection, build versus buy, reuse of existing accelerators versus bespoke work, and hosting approach.

• Write the technical sections of proposals and statements of work, and help the commercial team define outcome metrics that are measurable and defensible.

• Represent ConveneAI in client architecture, security, and vendor review boards, and clear the technical path to a signed deal.

Architecture Governance Across Pods

• Run architecture review at kickoff and at defined gates for every live engagement, and hold the gate when a design is not ready.

• Make the calls that carry risk beyond one pod: data residency, private versus hosted inference, identity and access design, network isolation, and integration patterns with systems of record.

• Unblock pods when they hit something outside their depth, by pairing and reviewing rather than taking the work away from them.

• Own the technical escalation path when an engagement drifts off track, and give the Product Owner and leadership an unvarnished read.

• Monitor inference cost, latency, and reliability across engagements, and intervene before an architecture choice erodes gross margin.

Platform, Standards and Reuse

• Own the ConveneAI reference architecture and accelerator library, and decide what gets harvested from an engagement versus rebuilt properly for reuse.

• Set the standards every pod inherits for agent design, evaluation, observability, security, and cost management.

• Evaluate new models, frameworks, and tooling with rigour rather than enthusiasm, and decide what enters the standard stack and what does not.

• Build the technical assets that let a four-person pod deliver what a fifteen-person team used to, because that gap is the entire commercial thesis.

Technical Leadership

• Mentor GenAI Engineers and Data Engineers across pods through design and code review, and raise the technical ceiling of the delivery organisation.

• Interview and calibrate technical hires, and help define the engineering career path as we scale from the first pods to many.

• Work with senior AI architects across our global operating bench to keep standards consistent across geographies.

WHAT WE'RE LOOKING FOR

• 10 or more years building and shipping enterprise software, including at least 2 years on production LLM or agentic systems that real users depended on.

• A track record as the technical owner across multiple concurrent client engagements, not just one long programme.

• Deep hands-on capability: Python, cloud architecture (AWS, Azure, or GCP), distributed systems, and API and integration design. You will still be expected to read and write code.

• Production experience with agent architectures, retrieval at enterprise scale, evaluation, and the operational side of AI systems: cost per task, latency, observability, and failure modes.

• Enterprise architecture fluency: identity and access, security posture, data residency, network isolation, and model hosting options, plus the experience to get a regulated client's review board to approve a design.

• Commercial literacy. You understand how an architecture decision flows through to gross margin on a fixed-fee engagement, and you make decisions accordingly.

• Credibility in both directions: with a client CIO or Chief Architect, and with an engineer who suspects you have stopped building.

• The judgement and standing to recommend against a deal that cannot be delivered, and to hold that position under commercial pressure.

• Legally entitled to work in Canada and willing to travel to client sites.

BONUS POINTS

• Architecture leadership at a consultancy or systems integrator, or a Field CTO or principal architect role at an AI, data, or cloud platform vendor.

• Depth in a regulated industry with named systems: policy administration and claims platforms such as Guidewire or Duck Creek, core banking, or clinical systems.

• Self-hosted or open-weight model deployment, fine-tuning, distillation, and private inference architecture.

• Experience building a reference architecture or accelerator library that other teams genuinely reused rather than ignored.

• Experience scaling a delivery organisation from a handful of teams to many, and the operational scar tissue that comes with it.

• Published writing, conference talks, or open-source work on applied AI architecture.

WHY JOIN CONVENEAI

• Your decisions compound. Architecture set here is inherited by every engagement that follows, which is a rare amount of leverage.

• You define the standard rather than inherit someone else's, at the point in the company's life when it actually gets set.

• Genuine breadth: pre-sales shaping through production operations, across multiple industries, without the politics of a large firm.

• Direct working relationship with the founders and with senior AI operators from BCG X, Infosys, Mu Sigma, and enterprise AI startups.

• Meaningful equity at the stage where technical decisions determine whether the delivery model scales or stalls.

Originally posted on Himalayas