Feedback Loop Engineering: making your project agent-ready

Explore feedback loop engineering, focusing on designing loops around agents to improve project conditions, enhancing autonomy and effectiveness in agentic coding.

Maria Gorinova, Simon Maple · 9 min read · 4 Sept 2026

What your design system can't teach AI agents

Tessl's software factory ships work no designer has touched, which exposed what a design system cannot hand to an agent: composition, the judgment of what goes on a screen and what to leave out. By having agents diff generated screens against hand-redesigned ones, the team turned that tacit judgment into checkable rules and shipped them as skills that guide agents at creation and review. The result is one designer's taste, encoded once and reaching every screen the factory produces.

Hamza Oza · 11 min read · 2 Sept 2026

Cheaper Tokens, Bigger Bills: Token Price Isn't Agent Cost

Token prices don't reflect the true cost of AI models in code reviews. Turns and verified outcomes are better measures, revealing hidden expenses in agent workloads.

Simon Rendon Arango · 13 min read · 28 Aug 2026

The Rise of the Harness Engineer

The rise of the harness engineer highlights a new role in managing AI-generated code, focusing on maintaining quality and addressing increased bugs and production incidents.

Amy Heineike · 17 min read · 25 Aug 2026

Vibe Coding Is Not The Best We Can Do

The article questions the reliance on AI coding assistants, emphasizing the importance of specification, verification, and tests in software engineering.

Dave Farley · 13 min read · 24 Aug 2026

New in Tessl Academy: Code Review Loops

Tessl Academy introduces Code Review Loops, a course on agentic code review, teaching skills like writing review lenses and handling pull requests in four concise lessons.

Alan Pope · 7 min read · 21 Aug 2026

Humans Architect The System, AI Writes The Code

The article explores a shift in software development where humans focus on system architecture while AI handles coding, emphasizing a new approach to creating reliable software.

Paul Stack · 11 min read · 19 Aug 2026

3 things you can do with Tessl Agent in your first week

Explore how Tessl Agent enhances coding efficiency by automating reviews, fixing recurring errors, and creating workflows, improving codebases without manual maintenance.

Alan Pope · 4 min read · 10 Aug 2026

Agents need real sandboxes

The article discusses the need for isolated environments, like Docker Sandboxes, to safely run AI coding agents, highlighting risks of agent autonomy on developer machines.

Oleg Šelajev · 15 min read · 30 Jul 2026

Agent Benchmarks Need To Measure The Whole Workflow

Agent benchmarks must evolve to measure entire workflows, as modern workloads involve complex, multi-turn interactions with tools, beyond simple chatbot exchanges.

Amit Kushwaha · 12 min read · 28 Jul 2026

What GitHub learned when better tools made Copilot code review worse

GitHub's migration of Copilot code review to shared tools initially worsened performance. Rewriting instructions improved accuracy and reduced costs by 20%.

Paul Sawers · 8 min read · 14 Jul 2026

Not all model ‘upgrades’ are upgrades — Microsoft data shows cheaper can cost more

Microsoft data reveals that newer AI models with lower per-token costs can end up more expensive due to higher token consumption, challenging assumptions about upgrades.

Paul Sawers · 7 min read · 7 Jul 2026