Founder & CEO @DynamiqAGI | Agentic AI
San Francisco, CA
- Tweets 346
- Following 629
- Followers 192
- Likes 214
Everyone is shipping code faster than they can read it.
Nobody reviews it. Nobody knows how it works. But it goes straight to production.
So @github decided to stop the madness by going down regularly.
@useblacksmith, @buildkite - are you paying GitHub for this marketing?
4
113
Train your own banking foundation model on @awscloud SageMaker. Inspired by @Revolut's PRAGMA paper, pragmatiq models banking unstructured data as timestamped event streams that can be turned into a useful embeddings.
We published a practical guide for running pragmatiq end-to-end on managed AWS infrastructure:
1. generate deterministic synthetic banking data, or bring your own parquet contract
2. prepare timestamped key-value event data in SageMaker Studio
3. train with a SageMaker @PyTorch training job
4. package the trained run for @nvidia Triton
5. serve user embeddings from a real-time AWS SageMaker endpoint
Link in the comments below 👇
2
2
173
It seems like @ManusAI got cut off from the frontier US LLM models and their performance degraded significantly over the past few weeks.
1
49
Today we're open-sourcing pragmatiq.
It's an independent implementation inspired by the PRAGMA paper: @Revolut's pre-trained banking foundation model.
What's inside:
- synthetic financial data generation
- tokenization + pretraining
- embeddings, probes, LoRA fine-tuning
- AML graph extension
- CPU-first pipeline with GPU acceleration
The goal is simple: give bank ML teams a clean, reproducible starting based on the original research paper.
Link in the comments 👇
1
1
170
LLMs are coming for average engineers first. And it's happening very fast.
They are not replacing great software engineers.
They are replacing the "engineers" who needed perfect tickets, five clarification meetings, and someone senior to explain the obvious. Someone who doesn't have critical thinking capabilities.
If your only skill is turning requirements into average code, bad news:
Claude Code or Cursor now does that faster, cheaper, and with no complaining.
Great engineers will be fine.
The rest are about to find out they were just expensive autocomplete.
1
3
238
MCPs are dead.
Not because AI agents do not need tools.
Because they do not need this new layer.
A good agent can already do the job.
It can write code.
It can call an API directly.
It can use a CLI.
CLI and direct APIs support progressive disclosure: the agent can start simple, inspect the results, and only ask for more when needed.
Turns out --help is a very good "protocol" on its own.
So what exactly is MCP adding?
In many cases: nothing.
Just one more wrapper. One more thing to maintain.
If an agent can already use APIs and CLIs, MCP is often redundant.
MCPs feel like a temporary phase that the industry will quietly move on from.
1
338
SaaS is dead.
We replaced all our subscriptions with "custom AI software".
Monthly spend dropped from $750 in SaaS to just $4570 in LLM tokens.
Big win.
As a bonus, the team now spends half their time fixing vibe-coded bugs instead of using working tools.
But at least we own the stack.
1
8
290
At some point you realize your agent doesn't need all that fancy stuff. It just needs SQLite.
Memory? SQLite.
Vector DB for RAG? SQLite.
Graph DB? SQLite.
Persistent state? SQLite.
System of record? SQLite.
Very advanced architecture, yes. One tiny file.
Let it sink.
1
3
182
Vitalii Duk retweeted
🗞️ We just released one of two case studies with @IBM. The piece shows how legal teams under pressure, from rising contract volumes and complex regulations, can move from reactive to proactive with Dynamiq’s legal agent.
Readers will see real outcomes:
➛ 50% faster contract review
➛ 98% faster answers to business queries
➛ 90% faster clause identification
You'll also find out how the solution uses IBM @IBMwatsonx.data, IBM Granite, and watsonx Orchestrate to plug insights into business systems.
🔎 Read the case study: ibm.com/case-studies/dynamiq
1
2
222
Vitalii Duk retweeted
📢 We’re live on @ProductHunt!
@prompt_signal = see your brand in AI answers + how to improve!📈
Would love your upvote & feedback 💚
👉producthunt.com/products/pro…
#llmvisibility #GEO #aimarketing
1
1
3
228
Vitalii Duk retweeted
Now live in @IBM Agent Catalog: @DynamiqAGI agents. Among the first partners in IBM Agent Connect, and first with more than one agent.
⛑️ Medical Research: structured clinical reports from trusted sources
cloud.ibm.com/catalog/servic…
👩⚖️ Legal Research: intake -> search -> structured memo/brief
cloud.ibm.com/catalog/servic…
Agent Catalog access, per-API pricing, IBM Cloud billing.
🎥 Integration guide: youtube.com/watch?v=wzUoTVod…
🎥 Medical agent guide: youtube.com/watch?v=w-cpEFEw…
🎥 Legal agent guide: youtube.com/watch?v=wzUoTVod…
2
2
4
163
Vitalii Duk retweeted
🚀 @prompt_signal hits @ProductHunt next Monday 6 October! See your brand in AI answers, and how to improve it. No PH pre-launch page, so watch this feed. We’ll share the link on launch morning. #ProductHunt #AI #marketing #GEO #AEO
1
4
141
Vitalii Duk retweeted
📢 Big update: you can now connect any #MCP server directly to agents in @DynamiqAGI! This means your agents can tap into custom tools, APIs, and data sources exposed via MCP, giving you more flexibility to design enterprise-grade AI workflows without workarounds.
2
3
112
Vitalii Duk retweeted
See the domains and pages LLMs rely on when they mention your brand. Sources helps you fix gaps and boost visibility. #LLM #SEO #llmvisibility
2
3
96
Vitalii Duk retweeted
We just published a new case study on how financial institutions can automate mortgage pre-approval using @DynamiqAGI together with @AmazonNova_ on @awscloud Bedrock.
In the article, we walk through how banks can design a 5-agent underwriting workflow (covering financial analysis, credit checks, property valuation, income verification, and senior risk review), enforce institution-specific policies at every decision point, and even plug in or mock Plaid/Experian data for testing.
You’ll also see how multi-agent orchestration with Nova improves explainability, speeds up reviews, and makes minutes-level pre-approvals possible without sacrificing compliance. If you’re curious about how AI agents are being applied to real financial workflows, this one’s worth a read.
👉 Read the case study: getdynamiq.ai/post/automatin…
1
4
106