AI Native Lab · Since Labs
0hours saved
0experiments validated

Since your team is still doing this by hand

Trusted by teams at
Boost
Founder Institute
Futuuri
Since AI
Mäkipelto

We test every AI tool on our own company first.
Only the ones that works reach yours.

The Lab

What we're testing right now

Every tool — and every workflow — goes through our own operations before we recommend it.

ToolWorkflowLLM model
New5
⚙ Workflow
Automated Competitor Intelligence
n8n→perplexity→email
⚙ Workflow
AI Sales Call Analysis
fireflies→claude→n8n
⚙ Workflow
AI Contract Review Workflow
claude→n8n→notion
⚙ Workflow
AI Vendor Due Diligence
claude→perplexity→n8n
⚙ Workflow
AI Employee Onboarding Q&A Bot
claude→notion→n8n
Up Next7
n8n mcp node
MCP tool integration for n8n agents
⚙ Workflow
RAG Internal Wiki Chatbot
notion→claude
⚙ Workflow
Site Photo → ISO-Compliant Daily Diary
claude→n8n→notion
Site photos → Claude vision extracts progress + safety notes → ISO-compliant daily diary entry
⚙ Workflow
Voice Memo → Structured Project Update
wispr→claude→notion
Voice memo → transcription → Claude structures into a standardised project update in Notion
⚙ Workflow
Municipal Budget Outreach Generator
perplexity→claude→n8n
Scans municipal budget publications → identifies relevant line items → drafts tailored outreach
⚙ Workflow
Automated Proposal Generator
claude→notion→n8n
⚙ Workflow
AI Customer Feedback Analysis
claude→n8n→notion
Testing6
Hermes
Open-source agentic QA & automation
Nous Research's self-improving agent — persistent memory, skills, browser automation. Evaluating for agentic test execution (reads a test plan, drives a real browser).
⚙ Workflow
Knowledge Base Auto-Update
claude→notion
⚙ Workflow
Hilma Tender Intel Brief
n8n→claude→notion
Pulls new Hilma procurement notices → Claude scores fit + drafts a go/no-go brief into Notion
⚙ Workflow
Inbox → Action Queue
claude→n8n→notion
Inbound email → Claude triage → structured action items pushed to a Notion task queue
⚙ Workflow
Meeting → Action Pipeline
notion→n8n→claude
All notes filed, action items created automatically — saves ~45 min/day
⚙ Workflow
Slack-to-Notion Knowledge Base Builder
claude→notion→n8n
Slack messages → n8n → Claude summary → Notion wiki entry, automated daily
Validated15
◎ Model
claude-sonnet-4-8
Anthropic · Reasoning & writing
Primary model for all Since Labs production workflows. Best instruction-following at scale.
◎ Model
claude-haiku-3-8
Anthropic · Fast & affordable LLM
Best cost-to-speed ratio for high-volume classification, routing, and extraction tasks
◎ Model
gemini-2.5-flash
Google · High-throughput multimodal LLM
1M context, native audio/video, 20x cheaper than Sonnet — best for high-volume pipelines
◎ Model
gpt-4o
OpenAI · Multimodal flagship
Strong vision capabilities; useful for eval comparison and multimodal pipelines
◎ Model
deepseek-r1
DeepSeek · Open-source reasoning LLM
Best open-source reasoning model — strong for math, code, and structured logic
◎ Model
deepseek-v3
DeepSeek · Open-source frontier LLM
Strong general-purpose open-source model at near-zero inference cost when self-hosted
ElevenLabs
Voice generation & TTS
Claude Code
Agentic coding CLI
Best for full-repo refactors, autonomous dev tasks, and MCP-integrated agent loops
n8n
Self-hosted workflow automation
Core orchestration layer for all Since Labs AI agent stacks. Cheapest at scale.
perplexity
AI-powered research & search
notion AI
Knowledge management
Good for wikis and quick lookups; use custom RAG for production agent workflows
notion AI meeting notes
AI meeting documentation
Validated for async meeting docs — not strong enough to replace dedicated transcription
gemini
AI assistant & research
Strong for long-context document analysis and multimodal tasks via the web interface
supabase pgvector
Vector database & backend
Preferred vector store for all Since Labs RAG pipelines — pgvector on managed Postgres
⚙ Workflow
RAG Document Q&A Pipeline
claude→supabase→voyage ai
Core Since Labs RAG template — PDF/URL → Firecrawl → Voyage embeddings → pgvector → Claude
Skip11
◎ Model
gpt-4.5
OpenAI · Multimodal LLM
Marginal gains over GPT-4o at significantly higher cost — not worth it for most workflows
◎ Model
llama-4-maverick
Meta · Open-source frontier LLM
Claude Sonnet wins on instruction following and tool use; use only if self-hosting is required
◎ Model
llama-4-scout
Meta · Efficient open-source LLM
Smaller Llama 4 variant — outclassed by Haiku 3.5 on instruction following
◎ Model
llama-3.1-405b
Meta · Large open-source LLM
Previous gen; superseded by Llama 4 — self-hosting cost not justified vs. API alternatives
◎ Model
mistral-small-3.1
Mistral · Efficient open-source LLM
Good but Claude Haiku wins on instruction following at similar price points
◎ Model
qwen-2.5-max
Alibaba · Open-source frontier LLM
Competitive but harder to deploy reliably vs. Claude or DeepSeek alternatives
◎ Model
phi-4
Microsoft · Small language model
Impressive for size but not a replacement for frontier models in production workflows
◎ Model
gemma-3-27b
Google · Open-source LLM
Good open-source option but Gemini 2.0 Flash API is cheaper and more capable
make.com
Workflow automation
Replaced Zapier for anything beyond 2-step flows
zapier
Simple trigger automations
Best for simple 2-step integrations only
OpenClaw
Open-source personal AI agent
Local-first messaging-channel agent (WhatsApp/Telegram/etc). Fun, but n8n + Claude covers our automation needs with better control and auditability.

26 tools evaluated — full notes, weekly updates, and our reasoning for every rejection.

View the full lab board →
Services

How we work together

One focused engagement to find where AI removes real work from your team — then the support to act on it.

01AI AuditFixed scope

“Walk away with a prioritized map of every hour AI can give back.”

We find exactly where AI can eliminate work in your company — then help you understand and act on what we found.

What's included
  • Full workflow analysis
  • Gap identification report
  • Step-by-step implementation plan
  • Live Q&A support
Why Since Labs

We don't recommend tools we haven't used.

The problem with AI consultants

Most AI agencies will show you a deck of 40 tools and tell you to "explore the possibilities." You leave with homework, not a plan.

We run every tool through our own company operations first — finance, client work, ops, hiring. If it doesn't hold up under real workloads, it doesn't reach your company.

01 — HOW WE FIND GAPS

Deep workflow analysis. We map every repetitive, high-volume task your team runs — looking for patterns, bottlenecks, and places where humans are doing what software should.

02 — WHAT YOU GET

A full audit report. Problems found, validated tools that solve them, and a step-by-step plan written for your specific stack and team size.

03 — HOW THE ENGAGEMENT WORKS

First, diagnosis. We map the gaps. Then, support. We answer every question about the report, adjust the plan, and unblock your team.

The Audit

What happens, step by step

01
Workflow discovery

We interview your team and map every repetitive, high-effort task. Nothing is off limits.

02
Gap analysis

We cross-reference your workflows against validated tools and identify exactly where time and money are being burned.

03
Report delivery

A full audit lands in your inbox: every gap found, the recommended tool, and an implementation plan your team can execute.

04
Implementation support

We're live to answer any question about the plan. Adjustments, clarifications, unblocking — whatever it takes.

We start whenever you're ready.

Start here

Ready to find out what's burning your team's time?

A full audit. A plan your team can actually execute.