Aleξξξ (@remotealex) header
Aleξξξ (@remotealex) profile picture

🇬🇷 sharing ai tips for less work / 🇪🇺

Joined July 2009
Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)Photo by Aleξξξ (@remotealex)
Aleξξξ retweeted
realizing all the hard shit you accomplished on your resume pre 2026 is just a prompt now
103
541
69
9,133
219,529
BTW anyone not using agents like this is falling behind and will inevitably wind up in the permanent underclass.
Asked one of our developers how he uses his AI agents and he just sent me this:
61
130
8
3,197
239,434
Aleξξξ retweeted
Julius convinced me to move from extra high to high on Opus 5.5 a week or two ago. He was right. This is better almost all of the time.
283
116
44
5,063
267,867
Long Greek AI talent
12
37
7
283
13,839
Aleξξξ retweeted
Life since Opus 5.5 dropped
18
94
5
1,540
39,395
So they pay you a $150K salary just to be a middleman between Claude and your boss?
414
2,016
232
63,879
1,617,280
🇫🇮Χαμός στη Φιλανδία που η Google έκοψε αυθαίρετα 400,000 δέντρα για να κατασκευάσει data center χωρίς να έχει ολοκληρωθεί περιβαλλοντολογική μελέτη. Το κράτος σταμάτησε άμεσα την κατασκευή. Στην Ελλάδα θα φωνάζαμε Α & ΟΥ & ΔΑΠ ΝΔΦΚ με την μίζα στην τσέπη teapawt.org/AFP/status/21125748698…
🇫🇮 Finnish town braces for change from massive Google data centre With its cool climate and relatively stable and cheap electricity, Finland has seen a spike in data centre investments. But as dozens of the energy guzzling centres are now under construction and many more planned by companies including Google, Microsoft and TikTok, uncertainty around energy prices, grid capacity and environmental effects have sparked a public debate in Finland
3
53
1
86
1,584
Europe's best climates, by how many days per year of truly nice weather you get. Canary Islands always FTW. I listened to your feedback, so this map is different: it rewards how many months sit in the 20–28°C window, and how often those days are sunny, regardless of peaks or excessive swings. In the end, great weather all year round isn't less good because one month is “too hot”, at least that's what I think...
Europe's best climates and where to find them, updated. Calabria FTW, Sicily not so much... Cyprus is surprisingly out because of very hot summers. Madeira and the Canary Islands remain near perfect, of course. This is the accurate data, with stricter parameters. Apologies for the mix-up: I was on mobile all day and crossed the order of the maps. I meant to publish a comparative 2000-2020-2050. Not too bad, but still, this is the one truly on point
74
40
19
643
304,440
Replying to @MistaksWerMade
What problem did this solved during those times.....?
This quoted post is unavailable.
81
24
6
1,359
31,288
#1 on github trending now...I do what I can😂
Reverse Engineer Anything 😂 Software is done.
109
451
22
10,295
593,389
Aleξξξ retweeted
this is what a european datacenter looks like
Replying to @GuillaumeLample
ML4 was trained on 3,800 NVIDIA Grace Blackwell GPUs in our European datacenters -- our cluster in Bruyères-le-Châtel built using our Series B fundraise. We are investing heavily in infrastructure, with our Series C and D clusters coming online soon. This additional compute will let us train longer, pursue more ambitious post-training, and iterate faster. We expect large and rapid improvements in the weeks and months ahead. 2/n
Readers added context they thought people might want to know
The posted photo shows the Breuillet - Bruyères-le-Châtel train station in France, not a datacenter. Mistral AI's cluster is in a separate modern modular facility operated by Eclairion at the location. fr.wikipedia.org/wiki/Gare_de_B… datacenterdynamics.com/en/news/mistra… usinenouvelle.com/article/ce-dat… baxtel.com/data-center/ec…
160
299
52
9,734
449,280
Aleξξξ retweeted
Prompt
99
52
93
1,724
1,109,031
Aleξξξ retweeted
I gave Claude another 18 hours.. and I think this one is the best one yet Macrohard: Windows XP I'm blown away
Gave Claude another 12 hours to one up this.. Woke up to.. The Clodyssey
762
1,753
735
17,060
2,053,794
Replying to @ClaudeDevs
Had Opus 5.5 make a video, insane. @bcherny @trq212 @lydiahallie @adocomplete @claudeai @AnthropicAI
4
3
1
97
25,228
Aleξξξ retweeted
My god this is such a good speech that every SWE needs to hear. You know what? Every person should hear it Keep the happy memories, eyes on the reality, be excited about the future. That’s the best that anyone can do
374
2,072
659
15,837
3,707,397
Me cold-calling my friends to play WoW Forever #worldofwarcraft #warcraft
38
209
27
3,183
254,994
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution
4,006
8,256
6,821
76,486
40,215,345
If you can build these 12 Forward Deployed Engineer projects. You're hired. Project 1 :) Bi-Directional Legacy Sync Engine Real-time sync between your modern SaaS and a customer's 20-year-old on-prem SQL database with conflict resolution. ➣ Shows : You can bridge modern AI with messy, real-world enterprise data. Project 2 :) Zero-Trust Multi-Tenant RAG Vector database with strict row-level security where Tenant A's AI can never see Tenant B's embeddings. ➣ Shows : You can deploy AI without leaking data across customer boundaries. Project 3 :) PII Redaction & Unmasking Proxy Middleware that intercepts LLM traffic, masks PII before it hits the model and unmasks it on the way back. ➣ Shows : You understand enterprise privacy and compliance (HIPAA/SOC2). Project 4 :) Air-Gapped One-Click Deployer Helm charts and scripts that deploy the entire AI stack into a secure, offline customer VPC in under 30 minutes. ➣ Shows : You can ship software into highly secure, restricted environments. Project 5 :) ROI & Usage Telemetry Dashboard A real-time dashboard showing the customer exactly how many hours and dollars the AI saved them this week. ➣ Shows : You prove product value to the customer's CFO not just their engineers. Project 6 :) Idempotent Webhook Reconciler An event processor that handles out-of-order webhooks, retries with exponential backoff and dead-letter queues. ➣ Shows : You build reliable integrations that survive network chaos. Project 7 :) Enterprise SSO & SCIM Bridge Automated user syncing between the customer's Okta/Azure AD and your app, handling role changes instantly. ➣ Shows : You can pass enterprise IT and security reviews. Project 8 :) AI Fallback & Degradation Gateway An API gateway that automatically falls back from Opus → Sonnet → Local Llama → Cache if the primary model times out. ➣ Shows : You protect the customer experience when upstream APIs fail. Project 9 :) Automated Compliance Auditor A script that continuously checks customer deployments against security controls and generates a readiness report. ➣ Shows : You treat security as code not a manual checklist. Project 10 :) Custom MCP Server for Legacy ERPs A Model Context Protocol server that lets an AI agent safely query and update a customer's SAP or Oracle instance. ➣ Shows : You can connect AI agents to the tools enterprises actually run on. Project 11 :) Shadow Traffic Evaluator A tool that duplicates 5% of production customer traffic to a new AI model silently, comparing outputs without affecting the user. ➣ Shows : You can safely test and upgrade models in production. Project 12 :) 3 AM Incident Runbook Repo A fully interactive, tested playbook repository that allows a junior engineer to rollback, scale or debug a customer outage at 3 AM. ➣ Shows : You build systems that others can recover when things break. Most people stay stuck watching tutorials. Builders get hired. bookmark & repost !
If I had 6 months to become a Forward Deployed Engineer. I'd do this. Stage 1: Python and API Foundations FastAPI, async, error handling, webhooks, REST/GraphQL, third-party SDKs. Stage 2: Customer Integration Patterns OAuth2, SAML, SSO, Salesforce/Slack/HubSpot APIs, webhook verification, retry logic. Stage 3: Multi-Tenancy and Data Isolation Row-level security, tenant identifiers, separate databases vs schemas, data partitioning. Stage 4: Security and Compliance Basics SOC 2, GDPR, HIPAA, PII redaction, encryption at rest and in transit, audit logging. Stage 5: RAG for Customer Data Secure ingestion pipelines, access control per document, citation tracking, permission filtering. Stage 6: Deployment Automation Docker, Kubernetes, Helm charts, CI/CD pipelines, one-click customer environments. Stage 7: Observability and Dashboards Distributed tracing, customer-facing health pages, usage metrics, SLA monitoring, alerting. Stage 8: Incident Management and Debugging Production debugging, log analysis, rollback strategies, post-mortems, customer communication during outages. Stage 9: Documentation and Runbooks Deployment guides, API docs, troubleshooting playbooks, onboarding checklists, knowledge base articles. Stage 10: Requirements Gathering and Translation Talking to non-technical stakeholders, translating business needs to technical specs, managing expectations. Stage 11: Customer Success and Adoption Usage analysis, feature adoption tracking, identifying expansion signals, reducing churn, proving ROI. Stage 12: Public Portfolio and Case Studies Document deployed projects, write implementation guides, share integration patterns, publish success metrics. Most people stay stuck watching tutorials. Builders get hired. (Bookmark it)
46
114
1
990
115,075
I wonder if there is an @xkcd comic for this...
1
7
187
13,109