Production AI engineering for startups

Build AI systems that work beyond the demo.

I help startup teams turn AI prototypes into reliable products. You work directly with me, from architecture through deployment.
Bryce Sol

Bryce Sol

AI engineer & founder · 8 years at startups

20 minutes · you leave with a clear next step

Fig. 1Matera · time to find an answer

By hand30 min

With the assistant5 min

100h/day

saved by property managers with an AI assistant

Matera case study →

Startups I've worked with

AlmaManoManoOpioMatera

“Bryce independently led an AI project from A to Z, now used daily by about a hundred people. This project turned out to be crucial for the company's future and convinced us to invest more in it.”

Hugo Lime

Hugo Lime

VP Data & AI at Matera

Matera case study →

Selected work

Shipped, in use, and measured.

Each runs on the client's own documents and data, and each is measured against a test set.

PropTech · AI Engineering

Matera: AI assistant searching 6M+ docs

Built an internal AI assistant from scratch: a chatbot on top of a knowledge search engine that retrieves the information employees need to do their work, across 6M+ documents.

Fig. 2Matera · questions asked per month since launch
×60

100h/daysaved across Matera's property managers

Financial Services · AI Engineering

Opio: Due diligence at scale

Rewrote the RAG system behind Opio's financial due-diligence platform from scratch, and worked across the product: the document ingestion pipeline, the full-stack app and automated deployments.

Fig. 3Opio · the client's own 60-question eval

Before · 30 of 60 right

After · 60 of 60 right

50% → 100%accuracy on the 60-question baseline eval

“I had the pleasure of working with Bryce for several months, and I can confidently say he's one of the most reliable and technically skilled developers I've worked with. [...] He was immediately operational and quickly became a key contributor across multiple areas: frontend, backend, agentic systems, and CI/CD. [...] I can't recommend him highly enough.”

Olivier Chance

Olivier Chance

CTO at Opio

Opio case study →

How I help

Where teams bring me in.

  1. Where to start

    Data & AI strategy

    You have more AI ideas than engineers. We pick the one with the fastest business impact, check the data can support it, and plan the first milestone.

    “…convinced us to invest more in it.”Hugo Lime, VP Data & AI at MateraMatera case study →
  2. What users see

    New AI feature

    You know which business workflow needs to scale. I scope it with you and build it.

    ~100Matera employees use the AI assistant I built every day to search 6M+ documentsMatera case study →
  3. What makes it reliable

    Prototype to production

    Your demo works, but real users break it. I measure where it fails, fix it, and ship it.

    50% → 100%accuracy on the 60-question baseline evalOpio case study →
  4. The foundation

    Data platform

    Your agents and dashboards are only as good as the data under them. I build or fix that foundation.

    24h → 5 mindata freshness for the application databaseAlma case study →

Not sure which layer you're on? That's what the call is for.

Book a free call

“Bryce worked with us for about a year to build a modern data stack from scratch at Alma. His expertise and rigor were invaluable during his mission and the whole team progressed tremendously thanks to him.”

Marin Huet

Marin Huet

Head of Data at Alma

Alma case study →

Data platforms and ML in production.

E-commerce · ML & Data

ManoMano: E⁠-⁠commerce optimization

Worked on the Google Shopping bidding algorithm and built the keyword-campaign system, alongside a delivery-time model, for a European e-commerce unicorn.

Fig. 4ManoMano · Shopping bidding model, randomized A/B test

Controlbaseline

New model · +$30M/yr+7%

+$40M/yradditional business volume from the bidding & marketing ML

Fintech · Data Platform

Alma: Data platform

Led the build of Alma's data platform from scratch. From cloud infra to business dashboards.

Fig. 5Alma · what the platform connects

Sources

What it feeds

  • Postgres
  • HubSpot
  • Zendesk
  • Stripe
  • Adyen
  • Google Sheets
BigQuery · dbt
  • Dashboards
  • Fraud model
  • HubSpot
  • Zendesk
  • Customer.io

Postgres via CDC: freshness 24 h → 5 min

$50M/yrfraud savings enabled by the platform

“Bryce successfully worked on complex and strategic data projects for the company. [...] Bryce has a rare multidisciplinary profile that gives him a precise understanding of the challenges of each data team (Engineering, Analytics, Science, MLOps, ...), and therefore lets him work with each of them.”

Romain Ayres

Romain Ayres

Head of Data Science at ManoMano

ManoMano case study →
Bryce Sol
Fig. 6Bryce Sol, AI engineer & founder.

Who you work with

Hi, I'm Bryce.

I'm an AI engineer with eight years in startups. I started in data science and data engineering, then moved into LLM systems, so I build AI from the data up.

Pyramind is my data and AI consultancy, and the name is the method. An AI product only holds up on strong foundations: clean data, layers that each do one job, and a data strategy that ties them together. When an AI project stalls, the problem is usually one layer down.

You work with me directly, from the first call to the handover, and your team ends up with a system it can run without me.

“One of our significant achievements together was building the data platform at Alma. Bryce's dedication and expertise were instrumental in its success. Their ability to communicate effectively and collaborate with the team ensured a smooth and efficient development process.”

Hamza Sefiane

Hamza Sefiane

Lead Data Engineer at Alma

Alma case study →

FAQ

Before you book the call.

Which startup projects are a fit?

The best projects have a concrete workflow, access to relevant data, and a team that will use the result. It is not a fit if there is no data yet, if the question is pure research, or if you need a full team rather than one engineer. If you are still deciding where AI fits, start with a free 20-minute call.

Book a free call →
We're hiring for this role. Why a freelancer?

Because a hire takes months, and the work can start now. I cover the seat in the meantime, then hand over to the person you hire. At Matera, I led an AI project from A to Z: about a hundred people now use it every day, and their VP Data & AI calls it crucial for the company's future. At Alma, the only data engineer was leaving: I led the build of the data platform, coached the engineer hired four months in, and handed it over to the team that grew around them. At Opio, I was the first engineer beside the CTO after the raise, and handed over once the internal team was ready.

Can you work with our existing team and stack?

Yes. I work alongside your engineers and product people, on the stack you already run. At Matera, the product is Ruby on Rails: I built the GenAI layer in Python beside it, and their Rails developers contributed to it, with me reviewing their changes. At Opio, I worked across the whole product, from the retrieval system to the React app and the deployments. I have shipped on GCP, Azure and AWS. Handoffs are documented, so your team can own the system at any point.

How will we work together?

We review your workflow, data, and integrations, then agree on scope, milestones, and a delivery timeline. We start with the smallest version that is safe to ship: read before write, batch before real time. Cost and risk go up only once it is proven. I work remotely and asynchronously inside your team, with a 30-minute weekly call to steer the project.

How do we know it will work?

Because the methodology has been battle-tested, with success, on a wide range of problems and industries. There is no secret sauce: it is what the top scale-ups do. Put a first version in front of users fast, then improve it in short iterations on what they actually do with it. What makes that possible is breadth: I cover data, agentic systems, web and infrastructure myself, so a project never waits on someone else to move forward. Reading about these methods in a blog post is not enough. It takes experience in the field to build the intuition that gets those results, and every project I have worked on in the last eight years has succeeded.

Our data is sensitive. How do you handle it?

We sign a contract with an NDA before the work starts. Everything runs on your infrastructure, so your data stays in your accounts, under your access rules.

You're one person. What if you're unavailable?

The project never depends on me alone. Everything runs on your infrastructure from the first day, work is documented as we go, and your team follows the work in the weekly call, so it can pick the system up at any point. A clean handoff takes about two weeks. You also get access to my network of vetted data and AI experts: one of them can step in within a week.

Who owns and operates the system afterward?

You do. The system ships with documentation and observability, so your team can run it without me. Staying on for ongoing improvement is a choice, not an assumption.

What does it cost?

It depends on the scope, so I don't publish a rate. I bill per day worked, and the price for the whole scope is fixed in the proposal before anything starts. The first call is free, and the audit is the smallest way to begin.

Book a free call →
What's the minimum commitment?

An audit of your proposed use case or existing system over a few weeks. You receive a prioritized plan covering the main blockers, what to build first, and a first milestone to validate before committing to a longer engagement.

Tell me what you're trying to ship.

Bring the workflow and what's blocking it. In 20 minutes, you'll know whether AI can solve it, what to build first, and whether I'm the right person for it.

20 minutes · you leave with a clear next step

Book a free call