AI Engineer
← All organizations

AI education and enterprise deployment

Towards AI

Towards AI provides AI education and builds custom AI systems for enterprises. Towards AI Academy offers project-based courses that take developers from data pipelines to deployment, while enterprise programs combine workforce training, architecture advice and hands-on development. Teams can commission documented LLM systems and learn to maintain them internally. Applications include receipt processing, document automation and tender discovery, with services extending to private equity portfolios.

The original platform launched in 2019 under founder Roberto Iriondo. Today, its leadership includes CEO and co-founder Louie Peters, CTO and co-founder Louis-François Bouchard, and co-founder Denis Piffaretti. Peters and Bouchard co-authored Building LLMs for Production, which teaches prompting, retrieval-augmented generation, fine-tuning and application reliability. Its public companion notebooks turn those techniques into practical exercises covering applications, evaluation, agents and inference benchmarking.

A 2022 acquisition brought the original platform’s assets into Towards AI, Inc., a newly incorporated company led by Peters that adopted the Towards AI name. By August 2026, the company reported 500,000+ learners taught since 2019, 200,000+ newsletter readers, 100,000+ Discord members and 10,000+ book copies sold. Its enterprise practice and educational programs are supported by a team of more than 15 AI practitioners.

towardsai.com ↗

3 talks

Newest first

3 speakers at AIE

Affiliations reflect their AIE appearances, not necessarily current employment.

Louis-François BouchardCTO & Co-FounderSamridhi VaidSenior Machine Learning EngineerOmar SolanoAI Engineer

Start here

  1. Build Your Own Deep Research Agent + Technical Writer

    Start here to learn how an MCP-oriented research agent connects grounded research and YouTube analysis to a more constrained technical writing workflow.

    Louis-François Bouchard · Paul Iusztin · Samridhi VaidAI Engineer Europe 2026

  2. Context Engineering in 2026: Compaction, Memory & Cost

    Learn how the presenters evaluate an AI tutor’s context, including Opik observability, time to first token, and the cost of evaluation runs.

    Louis-François Bouchard · Omar Solano · Samridhi VaidAI Engineer World's Fair 2026

  3. Turn 10,994 Notes Into Your Agents' Memory

    See how AI Research OS converts notes and other personal knowledge sources into persistent context for agents.

    Paul Iusztin · Louis-François BouchardAI Engineer World's Fair 2026

Messages from the stage

Evaluate writing beyond the prompt

The joint research and writing session discusses constrained writing workflows alongside LLM judges, held-out dataset splits, and precision, recall, and F1 scores to expose quality problems and overfitting.

Measure the tradeoffs of conversational context

The AI tutor workshop examines compaction and delta summarization alongside caching and full-history baselines, with attention to multi-turn recall, latency, and evaluation costs.

Make research memory reusable

The joint AI Research OS session discusses a file-based memory layer and lightweight personal research wikis, comparing one-off tools with reusable deep-research workflows.

Affiliations reflect each recorded session, not necessarily current employment. The research-and-writing and AI Research OS sessions are joint presentations with Paul Iusztin of Decoding AI.

Company sources · checked 2026-08-27